2 LIVERMORE 45-134 (DO NOT DELETE)
11/5/2015 9:20 AM
POLITICAL PARTIES AND
PRESIDENTIAL OVERSIGHT
Michael A. Livermore*
ABSTRACT.................................................................................................... 46
INTRODUCTION ............................................................................................ 46
I. PRESIDENTS, PARTIES, AND ADMINISTRATION ........................................ 53
A. Executive Restructuring............................................................. 54
B. Presidential Control in Administrative Law .............................. 58
C. Decline and Rebirth of American Political Parties ................... 62
II. THE MECHANICS OF PRESIDENTIAL OVERSIGHT .................................... 68
A. People Are Policy ...................................................................... 68
B. Parties and Appointments.......................................................... 72
C. Performance, Politics, and the New Patronage ........................ 76
III. PARTIES AND PRESIDENTIAL REPRESENTATION .................................... 80
A. A Simple Model of Presidential Representation ........................ 81
B. Politicians and Groups in Political Parties............................... 86
C. Responsible Party Government ................................................. 93
IV. RELATIONSHIPS WITH CONGRESS .......................................................... 96
A. Shared Oversight ....................................................................... 97
B. Party Discipline in Congress ................................................... 100
C. Interbranch Sparring ............................................................... 104
V. RESPONSIBLE PARTY ADMINISTRATION ............................................... 110
A. Administrative Values and Party Rivalry ................................ 111
B. Judicial Review ........................................................................ 117
C. Executive Review ..................................................................... 124
CONCLUSION .............................................................................................. 132
* Associate Professor, University of Virginia School of Law. My thanks to John Aldrich, Rachel
Barkow, Jessica Bulman-Pozen, Daniel Farber, Michael Gilbert, Sally Katzen, David Lewis, Sidney
Milkis, Lia Norton, Richard Pierce, Rachel Augustine Potter, Richard Revesz, Arden Rowell, Mark
Seidenfeld, Sidney Shapiro, Mark Tushnet, Adrien Vermeule, Melissa Wasserman, B. Dan Wood, and
to participants at workshops at Duke School of Law, the University of Illinois (Urbana-Champaign)
College of Law, the University of Virginia School of Law, and Vanderbilt Law School for valuable
comments. Thanks also to Caroline Stewart for excellent research assistance.
45
2 LIVERMORE 45-134 (DO NOT DELETE)
46
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
ABSTRACT
The intimate, and often underappreciated, relationship between
political parties, the presidency, and the administrative state has profound
consequences for the functioning of these three institutions at the heart of
American political life. Failure to draw out and illuminate this connection
has led to an incomplete understanding of administrative law—both
descriptively in what the law does and can do, and normatively in what it
should do. In particular, courts and commentators have not grappled with
the fundamental shift experienced by political parties over the past several
decades—from traditional party organizations that were locally focused
and ideologically diverse to today’s high-tech, polarized, and nationally
oriented parties. This transformation was strongly influenced by the growth
of administrative agencies and the modern presidency, and has important
implications for the practical realities of presidential control over
agencies, for the normative justifications supporting the President’s
prominent role in the administrative state, and for the struggle over control
of agencies between the President and Congress. This Article maps out the
intersection of President, administration, and contemporary parties and
proposes “responsible party administration” as a normative framework to
evaluate how well institutional arrangements and judicial doctrines
forward administrative values such as expertise, coherence, and legality
while accommodating a system of regulated rivalry in which two
programmatically distinct parties compete for the reins of the
administrative state.
INTRODUCTION
In the past several decades, the intimate relationship between political
parties, the presidency, and the administrative state has undergone a basic
transformation. Traditional political parties—dominant for nearly a century
and caricatured in depictions of smoke-filled backroom deals between party
bosses—effectively disappeared.1 Into that void stepped a “new breed” of
contemporary political parties that are primarily concerned with policy
programs, rather than patronage, and that are nationally oriented, rather
1. See infra Part I.C (discussing heyday and decline of traditional party organizations). See
RICHARD M. HARNETT & BILLY G. FERGUSON, UNIPRESS: UNITED PRESS INTERNATIONAL COVERING
THE 20TH CENTURY 48 (attributing first usage of “smoke-filled room” to describe political decision
making to journalist Raymond Clapper’s coverage of the 1920 national Republican convention, in
which successive failed balloting rounds led to a lengthy discussion between party leaders that resulted
in a compromise nominee, Warren G. Harding).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
47
than locally focused.2 The decline and revival of American political parties
has pervasively influenced the relationship between the President and the
administrative state.3 Although political scientists who study the
presidency, government bureaucracy, and political parties have examined
these changes,4 administrative law scholarship has not adequately grappled
with the normative and descriptive consequences of this profound shift.
Drawing on a substantial body of recent political science literature, this
Article examines the changed relationship between presidents, parties, and
administration in light of several central questions in administrative law.
The new breed of political parties has important and underappreciated
consequences for how presidents exert control over agencies, whether the
President’s prominent role in the administrative state is justified, and how
Congress and the President interact in their shared oversight roles. Perhaps
most important, contemporary parties contribute to broader trends that have
consigned traditional normative models of administrative law to a “lost
world” that is unlikely to return.5 New normative models are needed, better
capable of responding to current circumstances. In a spirit of exploration of
this new terrain, the Article proposes responsible party administration as
an alternative normative aspiration for administrative law. This framework
seeks to reconcile the benefits of partisan rivalry with administrative values
such as expertise, coherence, vigor, and legality. Striking this balance is no
simple task and existing institutional arrangements and legal doctrines
succeed to varying degrees under this rubric.
The transformation to contemporary parties gathered steam in the
decades leading up to the 1960s, when social, cultural, legal, and
technological changes undermined the mass political parties that had
structured political life in the United States for nearly a century and a half.6
The waning of traditional party organizations did not spell an end to
political parties. Instead, after a period in which parties “hovered in the
2. JEANE KIRKPATRICK, THE NEW PRESIDENTIAL ELITE: MEN AND WOMEN IN NATIONAL
POLITICS 3 (1976) (referring to “the ‘new breed’ hypothesis” that “American politics is being
transformed in some important, fairly fundamental ways by the ascendancy [in party politics] of large
numbers of new men and women whose motives, goals, ideals, ideas, and patterns of organizational
behavior are different from those of the people who have dominated American politics in the past”); see
infra Part I.C.
3. For a general discussion of the relationship between presidents and agencies, see Elena Kagan,
Presidential Administration, 114 HARV. L. REV. 2245 (2001).
4. See infra Part I.C. See Kenneth S. Lowande & Sidney M. Milkis, “We Can’t Wait”: Barack
Obama, Partisan Polarization and the Administrative Presidency, 12 THE FORUM 3 (2014) (examining
“the tension between presidential leadership, the administrative state, and modern political parties” in
the context of high profile executive actions of the Obama administration).
5. Daniel Farber & Anne Joseph O’Connell, The Lost World of Administrative Law, 92 TEXAS L.
REV. 1137 (2014).
6. JOHN H. ALDRICH, WHY PARTIES?: A SECOND LOOK (2011).
2 LIVERMORE 45-134 (DO NOT DELETE)
48
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
background,” they reemerged in the late 1970s and 1980s with renewed
institutional forms suited to the contemporary needs of office seekers,
interest groups, and policy-oriented activists.7
This transformation is sufficiently striking that it is possible to speak of
two distinct types of American political parties that were dominant during
different historical periods.8 Traditional party organizations, which had a
long heyday that lasted from the mid-nineteenth century through roughly
1960, were locally oriented, ideologically diverse, and motivated primarily
by patronage rather than policy.9 The contemporary political parties that
have supplanted this earlier form are nationally oriented networks of
affiliated interests that are highly professionalized and characterized by a
set of shared policy goals, rather than simply a desire for patronage jobs.10
This Article argues that the revival of American political parties has
had a ubiquitous and underappreciated influence on presidential oversight
of the administrative agencies.11 Without accounting for this influence, it is
impossible to understand many important features of the contemporary
administrative state. For example, as a purely descriptive matter, scholars
have failed to adequately explore how contemporary parties affect the
ability of the President to translate theoretical power into reality. Presidents
are thought to have two primary strategies for exerting control over the
7. Id. at 316 (citing Kenneth A. Shepsle, The Changing Textbook Congress, in CAN THE
GOVERNMENT GOVERN? 238 (John E. Chubb & Paul E. Peterson eds., 1989)).
8. The terminology introduced in this paragraph creates a rough distinction to categorize a
change in party structure in the second half of the twentieth century. See infra Part I.C. Of course,
different or finer-grained distinctions are possible.
9. There was considerable variation during a time period that began with the telegraph and
railroads and ended with the integrated circuit and air travel, but there was enough consistency in the
structure and purposes of political parties that Professor Aldrich has stated that “the basic structure of
the modern mass political party was reasonably well established by the 1860s [and] this form remained
intact until about 1960.” ALDRICH, supra note 6, at 163–64.
10. See MARTY COHEN, DAVID KAROL, HANS NOEL & JOHN ZALLER, THE PARTY DECIDES:
PRESIDENTIAL NOMINATIONS BEFORE AND AFTER REFORM 16 (2008) (arguing that the “mixed
coalition of party loyalists” that came to prominence over this period is “a real political party . . . rather
than a haphazard collection of special interests and unsavory characters”). This coalition includes
“officeholders, ideologues, fund-raisers, interest groups, and others.” Id.; see also ALDRICH, supra note
6, at 285 (describing campaign personnel—“pollsters, media advisers, direct mail specialists, campaign
finance and law experts, and all the rest”—as essential components of contemporary parties). Joseph
Fishkin and Heather Gerken use the term “shadow parties” to refer to this similar phenomenon. See
Joseph Fishkin & Heather K. Gerken, The Party’s Over: McCutcheon, Shadow Parties, and the Future
of the Party System, 2014 SUP. CT. REV. 175.
11. A few words on terminology may be helpful. The phrases “political party” and “partisan”
have multiple meanings that can be a source of confusion. Throughout this Article, the phrases
“traditional party organization” and “contemporary political party” will refer to the two distinct forms
of party organization discussed in the previous paragraph. “Political parties” will refer to both
traditional party organizations and contemporary parties. “Partisan” will be used throughout in its
adjectival sense to describe motives or actions that arise from political parties and their programs.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
49
executive: centralization and politicization.12 Centralization involves the
accumulation of authority directly in the White House and its offices.
Politicization involves the use of the appointments power to place
individuals who can be counted on to promote the President’s agenda in
senior management positions at agencies.
These strategies, and their consequences for administrative law and
regulatory outcomes, have been extensively discussed in the legal
literature.13 But there has been less attention paid to how presidents respond
to the challenge of identifying the cadre of loyal and competent personnel
who are needed to carry out these two strategies. “People are policy,”14 and
with the right information and few external constraints over personnel,
presidents can use centralization and politicization strategies to build an
“institutional presidency” that exerts powerful control over the
administrative state.15 This Article contributes to the literature on
presidential control by illuminating how the transition to contemporary
political parties helps explain the ability of presidents to identify the loyal,
competent personnel necessary to achieve policy goals through personnel
decisions.
In addition to helping to explain how presidential control is executed,
contemporary parties deeply affect the normative justifications for the
President’s influence over agencies. Defenders of presidential oversight
consistently argue that the President has a special relationship with the
electorate, and the Supreme Court frequently references the President’s
representative capacity in cases that enhance the presidency’s supervisory
power over agencies.16 Critics of enhanced presidential power over the
administrative state, meanwhile, question whether the President can be
12. See Terry M. Moe, The Politicized Presidency, in THE NEW DIRECTION IN AMERICAN
POLITICS 235, 235 (John E. Chubb & Paul E. Peterson eds., 1985). Both of these strategies are related
to the President’s supervisory role within the executive. There is a constitutional law debate concerning
whether presidents also enjoy “directive” power over agencies, although the practical significance of
this question may not be substantial. See generally Nina A. Mendelson, Another Word on the
President’s Statutory Authority over Agency Action, 79 FORDHAM L. REV. 2455 (2011) (discussing the
debate over “directive” authority).
13. See, e.g., Kagan, supra note 3, at 2376–85.
14. James P. Pfiffner, Presidential Transitions, in THE OXFORD HANDBOOK OF THE AMERICAN
PRESIDENCY 85, 94 (George C. Edwards III & William G. Howell eds., 2009) (internal quotation marks
omitted). This expression has been attributed to President Ronald Reagan’s transition. See id.; see also
Paul Laxalt & John Lofton, 60-Second Debate, N.Y. TIMES, Mar. 18, 1982, at B14.
15. See JOHN P. BURKE, THE INSTITUTIONAL PRESIDENCY: ORGANIZING AND MANAGING THE
WHITE HOUSE FROM FDR TO CLINTON (2d ed. 2000). The term “institutional presidency” typically
refers only to the White House, see id. at 6, although it is not clear that the political appointees in the
White House and at agencies have a fundamentally different relationship with the President.
16. See infra Part I.B.
2 LIVERMORE 45-134 (DO NOT DELETE)
50
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
counted on to genuinely promote majority preferences through agency
oversight.17
Political scientists who study contemporary political parties and
presidential elections are divided between a “politician-centered” camp that
views parties as subservient to the needs of politicians, and a “groupcentered” camp that believes that parties are mechanisms for organized
interest to “capture and use government for their particular goals.”18 A
politician-centered theory predicts that presidents will seek to maximize
electoral prospects by attuning oversight to the preferences of median
voters.19 A group-centered theory, however, predicts that presidential
influence will skew agency decision making toward special interests.20
The presidential representation hypothesis is inconsistent with the
reality of contemporary political parties, which at least to some extent are
driven by activists to depart from the preferences of median voters. This
Article argues, however, that an alternative normative justification for
presidential control is possible that is more consistent with the reality of
contemporary parties. Grounded in the concept of “responsible party
government” from the political science literature,21 this justification is
based on the ability of presidential control to help facilitate a party system
in which the major parties develop competing policy visions, test those
visions for voter appeal during elections, and implement them once in
office.22 Although neither party platform tracks the median voter, party
competition promotes vitality in the electoral process and innovation in
policymaking.23
Finally, although several scholars have noted that struggles between the
President and Congress over the reins of administrative agencies have
become increasingly heated in recent years,24 the potential for
contemporary parties to enhance democratic accountability has led to a
misdiagnosis of the current state of inter-branch relations. The rise of
contemporary parties raised a new set of questions about the role of parties
in structuring legislative politics and the interaction between Congress and
the President over agency oversight. An important new feature of this
dynamic has been the incorporation of regulatory policy into national
17. See, e.g., Aziz Z. Huq, Removal as a Political Question, 65 STAN. L. REV. 1, 63–66 (2013).
18. Kathleen Bawn, Martin Cohen, David Karol, Seth Masket, Hans Noel & John Zaller, A
Theory of Political Parties: Groups, Policy Demands and Nominations in American Politics, 10 PERSP.
ON POL. 571, 571 (2012).
19. See infra Part II.
20. See infra Part II.
21. See infra Part III.C.
22. See infra Part III.C.
23. See infra Part III.
24. See infra Part IV.A.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
51
politics. Commentators observing this change have argued that
congressional oversight is either abandoned during periods of united
government or used to engage in “blood-sport confrontations over agency
rulemaking” during periods of divided government.25 Both have costs. A
lack of congressional oversight reduces democratic accountability, while
“blood-sport” proxy battles impose resource costs on agencies and increase
the contentiousness and length of the confirmation process.26
But it is a mistake to allow these costs to obscure the democratic
potential created by the incorporation of regulatory policy into national
party politics. Congressional oversight of agencies creates opportunities for
the party not in the White House to criticize the administration and develop
and publicize alternative positions, which can be evaluated by voters during
elections. This process shares similarities with confrontations between the
cabinet and “shadow government” in parliamentary systems that can
facilitate democratic responsiveness.27 Although enhanced tensions
between the branches brought on by contemporary parties create costs for
the administrative state—in terms of delay and potential for gridlock—the
importance of those costs cannot be accurately assessed without
acknowledging the potential benefits of party competition over regulatory
policy.
Just as the rise of contemporary political parties may clash with the
constitutional system of separated power, the role of parties in structuring
oversight of agencies may also conflict with core values at the heart of the
administrative state, including neutral competence, representativeness,
coherence, and vigor. Extending the concept of responsible party
government to the agency context, the Article proposes responsible party
administration as a normative framework to evaluate how well law and
legal institutions balance the benefits of party government with the
requirements of sound administrative decision making. The pervasive
influence of contemporary parties on the administrative state gives rise to a
host of arrangements—from judicial doctrines to agency structure—that
can be evaluated according to the responsible party administration
paradigm. This Article explores two: judicial doctrines concerning
25. Thomas O. McGarity, Administrative Law as Blood Sport: Policy Erosion in a Highly
Partisan Age, 61 DUKE L.J. 1671, 1711–12 (2012) (quoting Neal Devins, Party Polarization and
Congressional Committee Consideration of Constitutional Questions, 105 NW. U. L. REV. 737, 758
(2011)); see also Kagan, supra note 3, at 2346–49 (arguing that opposite-party members of Congress
face incentives to focus oversight activities on regulatory matters that have been signaled to be high
priorities for the President).
26. See McGarity, supra note 25, at 1711–16. See also Anne Joseph O’Connell, Shortening
Agency and Judicial Vacancies through Filibuster Reform? An Examination of Confirmation Rates and
Delays from 1981 to 2014, 64 DUKE L. J. 1645 (2015).
27. See infra Part IV.C.
2 LIVERMORE 45-134 (DO NOT DELETE)
52
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
deference to agency decisions, and executive regulatory review carried out
by the Office of Information and Regulatory Affairs (OIRA) in the White
House.
This Article will proceed in five Parts. Part I provides a brief overview
of three interrelated trends that serve as a backdrop for presidential
oversight of agencies. The first is the reorganization of the Executive
Branch to bring agencies under more centralized control; important
landmarks include the Executive Reorganization Act (adopted under
President Franklin Roosevelt) that created the Executive Office of the
President and President Reagan’s Executive Order 12,291 establishing the
practice of centralized regulatory review. The second trend is the
development of a president-centric administrative law that has, to some
degree, insulated presidential influence over agencies from congressional
and judicial interference. The third is the decline of traditional party
organizations and the rise of contemporary parties.
Part II examines how contemporary political parties affect the
mechanisms of presidential oversight. First, it explains the claim that the
President can exercise effective oversight of the administrative state
through centralization and politicization. It then argues that contemporary
parties can help draw agency policy toward presidential preferences by
providing presidents with access to a pool of loyal and competent
professionals and giving the President relative freedom in settling
intraparty disputes over policy through personnel choices.
Part III reviews the role that majoritarian justifications have played in
legitimizing the extension of political oversight by the President, both by
courts and by commentators, and then evaluates that justification in light of
the structure and influence of contemporary political parties. It will
introduce the competing politician-centered and group-centered models of
presidential parties and describe the consequences of those models for
presidential oversight of agencies. It will also discuss the potential for such
oversight, even if not reflective of median-voter preferences, to promote
democratic accountability through responsible party government.
Part IV examines the role of contemporary parties in structuring the
relationship between congressional and presidential agency oversight. The
resurgence of strong legislative parties, organized around national policy
programs, appears to have created conditions ripe for particularly
contentious relationships between the executive and the legislature during
periods of divided government.28 Many commentators have responded to
28. Professor Cass Sunstein uses the word “partyism,” which he describes as a bias akin to
racism, to describe the current climate. See Cass R. Sunstein, Partyism, U. CHI. LEGAL F. (forthcoming
2015).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
53
this development with worries about gridlock, incivility, and obstruction.29
Although these concerns may have substantial merit, there are
countervailing benefits associated with the public development and
exposition of party programs on regulatory policy that ought to be weighed
against these costs.
Part V proposes and develops the concept of responsible party
administration as a normative framework for balancing the benefits of
vigorous party competition with administrative law values such as
rationality, consistency, expertise, and impartiality. Administrative values
may favor agencies that are relatively insulated from political oversight, but
removing agencies from political influence cuts policymaking off from
parties and the benefits they provide. To be successful, modern
administrative law must strike a middle course between agencies that are
too responsive to demands from extreme constituencies within political
parties and agencies that are too far removed from everyday democratic
politics.
I. PRESIDENTS, PARTIES, AND ADMINISTRATION
There are many actors who vie for control of the federal bureaucracy,
including presidents, Congress, courts, trade associations, issue advocacy
organizations, and bureaucrats.30 Although “no single entity has emerged
finally triumphant” in this contest,31 many observers have concluded that,
of all of the combatants, the President now sits in a uniquely powerful
position to influence agency decisions.32
The role of the President in the administrative state as it stands today is
the consequence of three mutually reinforcing trends over the course of the
past century: the internal restructuring of the Executive Branch to empower
centralized managerial control; the reform of administrative law to remove
or weaken judicial and congressional constraints on agencies and
presidential oversight; and the reorganization of American politics around
national, presidentially oriented, programmatic parties. The importance of
the first two trends for presidential oversight is well recognized. To provide
29. See infra Part IV.C. For an example, see MICKEY EDWARDS, THE PARTIES VERSUS THE
PEOPLE: HOW TO TURN REPUBLICANS AND DEMOCRATS INTO AMERICANS (2012).
30. Terry M. Moe, Control and Feedback in Economic Regulation: The Case of the NLRB, 79
AM. POL. SCI. REV. 1094, 1095 (1985).
31. Kagan, supra note 3, at 2246.
32. See, e.g., Moe, supra note 12, at 266–67.
2 LIVERMORE 45-134 (DO NOT DELETE)
54
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
context for this Article’s exploration of the importance of the third trend,
this Part offers a brief overview.33
A. Executive Restructuring
Since at least the New Deal, presidents have “engaged in a series of
strenuous efforts to assert control over administrative decision making.”34
These efforts respond to the growth of “performance-based expectations”
for the President and the expansion of administrative agencies into nearly
every corner of economic life.35 At the level of constitutional theory, there
is a question of whether presidents do, or ought to, have the power to direct
the actions of executive officials.36 At a more practical level, scholars have
characterized two control strategies that presidents have deployed:
centralization and politicization.37
Centralization involves the accumulation of authority directly in the
White House and its offices.38 Over the past several decades, presidents
have expanded the authority of White House offices over the budget,
national security, communications, regulatory policy, personnel, and
legislative affairs.39 They have effectuated this shift through executive
order, formal reorganization, and informal shifts in bureaucratic turf or
lines of responsibility.40
33. PAUL PIERSON, POLITICS IN TIME: HISTORY, INSTITUTIONS AND SOCIAL ANALYSIS 2 (2004)
(“Placing politics in time can greatly enrich both the explanations we offer for social outcomes of
interest, and the very outcomes that we identify as worth explaining.”).
34. Michael A. Livermore & Richard L. Revesz, Regulatory Review, Capture, and Agency
Inaction, 101 GEO. L.J. 1337, 1344 (2013).
35. Dennis M. Simon, Public Expectations of the President, in THE OXFORD HANDBOOK OF THE
AMERICAN PRESIDENCY 135, 145–46 (George C. Edwards III & William G. Howell eds., 2009)
(discussing the historical origins of contemporary expectations). Mashaw notes that the expansion of
the regulatory state was gradual. See JERRY L. MASHAW, CREATING THE ADMINISTRATIVE
CONSTITUTION: THE LOST ONE HUNDRED YEARS OF AMERICAN ADMINISTRATIVE LAW (2012). But
there is little doubt that it picked up speed with the New Deal agencies, such as the National Labor
Relations Board and the Securities and Exchange Commission, and then again with the risk-regulation
agencies, such as the Environmental Protection Agency and the Occupational Safety and Health
Administration. See generally RICHARD P. NATHAN, THE ADMINISTRATIVE PRESIDENCY 34–42 (1983)
(discussing the growth of counter-bureaucracy in the White House during the Nixon administration).
36. There is a large literature on this question. See, e.g., STEVEN G. CALABRESI & CHRISTOPHER
S. YOO, THE UNITARY EXECUTIVE: PRESIDENTIAL POWER FROM WASHINGTON TO BUSH 24–25 (2008);
Robert V. Percival, Presidential Management of the Administrative State: The Not-So-Unitary
Executive, 51 DUKE L.J. 963, 999 (2001).
37. Moe, supra note 12.
38. Id. at 244.
39. See generally MATTHEW J. DICKINSON, BITTER HARVEST: FDR, PRESIDENTIAL POWER AND
THE GROWTH OF THE PRESIDENTIAL BRANCH (1996) (discussing management style in the Roosevelt
administration and contrasting it with subsequent presidents, focusing in particular on the role of the
White House bureaucracy).
40. See generally NATHAN, supra note 35.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
55
Centralization has been accompanied by an explosion in the size and
sophistication of the presidential bureaucracy. The Executive Office of the
President (EOP) was created during the FDR administration through the
Reorganization Act of 1939.41 The EOP originally included the Bureau of
the Budget (the predecessor to the Office of Management and Budget) as
well as the White House Office.42 Over the years, the White House Office
has been transformed from a “few informal, generalist aides to FDR” to a
“highly specialized” bureaucracy with “more than 400 staffers.”43 The EOP
also houses the Council of Economic Advisers, the Council on
Environmental Quality, the National Security Council, the Office of
Science and Technology Policy, the Office of the U.S. Trade
Representative, and the Domestic Policy Council.44 Nelson Polsby has
characterized this bureaucracy as a “presidential branch”—distinct from the
Executive Branch—that “imperfectly attempts to coordinate both the
executive and legislative branches on its own behalf.”45
The presidential bureaucracy exercises both informal and formal
influence over agency decision making. Informally, White House officials
manage access to the President, develop policy initiatives, control
communications with the media and the legislature, and facilitate
interagency coordination.46 The archetypical example of formal influence
over agency action is executive review of regulatory decision making.
President Ronald Reagan’s Executive Order 12,29147 established
41. Reorganization Act of 1939, ch. 36, 53 Stat. 561–565 (1939). The Act was spurred by the
recommendations of the Brownlow Committee, which suggested increased presidential control over the
administrative state.
42. See Reorganization Plan No. 1 of 1939, 3 C.F.R. 248 (1939), reprinted in 5 U.S.C. app. at
582 (2012), and in 53 Stat. 1423 (1939).
43. JAMES P. PFIFFNER, THE MODERN PRESIDENCY 3 (6th ed. 2011). The White House Office
now includes the National Economic Council, the Office of Communications, the Office of Legislative
Affairs, the Office of Presidential Personnel, the Office of Public Engagement and Intergovernmental
Affairs, and the White House Counsel.
44. Executive Office of the President, THE WHITE HOUSE, https://www.whitehouse.gov/
administration/eop (last visited Sept. 25, 2015).
45. Nelson W. Polsby, Some Landmarks in Modern Presidential-Congressional Relations, in
BOTH ENDS OF THE AVENUE: THE PRESIDENCY, THE EXECUTIVE BRANCH, AND CONGRESS IN THE
1980S 1, at 20 (Anthony King ed., 1983).
46. Kagan, supra note 3, at 2299.
47. Exec. Order No. 12,291, 3 C.F.R. 127 (1982) reprinted in 5 U.S.C. § 601 at 473–76 (1988),
revoked by Exec. Order No. 12,866, 3 C.F.R. 638 (1994), reprinted in 5 U.S.C. § 601 at 638–42 (2000).
There were important predecessors to the Reagan Executive Order. President Nixon initiated a “Quality
of Life review,” which required the Environmental Protection Agency and the Occupational Safety and
Health Administration to engage in an interagency consultation process and provide an estimate of the
costs of proposed regulation along with a set of alternatives. THOMAS O. MCGARITY, REINVENTING
RATIONALITY: THE ROLE OF REGULATORY ANALYSIS IN THE FEDERAL BUREAUCRACY 18 (1991).
President Gerald Ford created the Council on Wage and Price Stability (CWPS) in the White House,
which exercised increased central control over agency rulemaking. Curtis W. Copeland, The Role of the
Office of Information and Regulatory Affairs in Federal Rulemaking, 33 FORDHAM URB. L.J. 1257,
2 LIVERMORE 45-134 (DO NOT DELETE)
56
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
centralized review over rulemaking in the newly created Office of
Information and Regulatory Affairs (OIRA), an entity in the Office of
Management and Budget. Under Reagan’s order, all major regulations were
required to be cleared by officials at OIRA before they could be adopted.
Centralized review along the model established by Reagan has continued
through subsequent administrations, with each President offering revisions
but leaving the essential structure in place.48 Debates about the normative
desirability of regulatory review have continued as well.49
The other strategy presidents use to assert control over agencies,
commonly referred to as politicization, involves “an attempt to infiltrate the
agencies through aggressive use of the appointment power.”50 To effectuate
a politicization strategy, presidents can increase the total number of
officials who are political appointees rather than civil servants and exert
greater control over appointments to ensure that they are filled by
individuals with a high degree of loyalty.51 The two major expansions of
political appointees occurred during the presidencies of Dwight D.
Eisenhower and Jimmy Carter.52 President Eisenhower created the category
of Schedule C—“positions of a confidential or policy determining
character”—in 1954 to be filled by political appointees outside the civil
1264 (2006). President Jimmy Carter went even further with Executive Order 12,044, which required
the newly created Regulatory Analysis Review Group within the CWPS to perform an economic
analysis for any major regulation. Exec. Order No. 12,044, 3 C.F.R. 152 (1979).
48. See Exec. Order No. 12,866, 3 C.F.R. 638 (1993), reprinted in 5 U.S.C. § 601 at 802 (2012);
Exec. Order No. 13,563, § 1(a), 3 C.F.R. 215 (2011), reprinted in 5 U.S.C. § 601 at 816 (2012);
Michael A. Livermore, A Brief Comment on “Humanizing Cost-Benefit Analysis”, 2 EUR. J. RISK REG.
13 (2011); Michael A. Livermore & Richard L. Revesz, Retaking Rationality Two Years Later, 48
HOUS. L. REV. 1, 12–19 (2011); Richard H. Pildes & Cass R. Sunstein, Reinventing the Regulatory
State, 62 U. CHI. L. REV. 1, 5–6 (1995).
49. See Lisa Schultz Bressman & Michael P. Vandenbergh, Inside the Administrative State: A
Critical Look at the Practice of Presidential Control, 105 MICH. L. REV. 47, 49–50, 75 (2006); Joseph
Cooper & William F. West, Presidential Power and Republican Government: The Theory and Practice
of OMB Review of Agency Rules, 50 J. POL. 864, 882–83 (1988); Christopher C. DeMuth & Douglas H.
Ginsburg, White House Review of Agency Rulemaking, 99 HARV. L. REV. 1075, 1082 (1986); Mark
Seidenfeld, A Big Picture Approach to Presidential Influence on Agency Policy-Making, 80 IOWA L.
REV. 1, 4 (1994); Sidney A. Shapiro & Ronald F. Wright, The Future of the Administrative Presidency:
Turning Administrative Law Inside-Out, 65 U. MIAMI L. REV. 577, 613 (2011); RENA STEINZOR ET AL.,
CTR. FOR PROGRESSIVE REFORM, WHITE PAPER 909, A RETURN TO COMMON SENSE: PROTECTING
HEALTH, SAFETY, AND THE ENVIRONMENT THROUGH ‘PRAGMATIC REGULATORY IMPACT ANALYSIS’ 2
(2009).
50. Kagan, supra note 3, at 2277 n.123.
51.
See Andrew Rudalevige & David E. Lewis, Parsing the Politicized Presidency:
Centralization and Politicization as Presidential Strategies for Bureaucratic Control 4 (Sept. 1, 2005)
(unpublished manuscript) (on file with the Alabama Law Review) (suggesting that politicization also
includes “involving civil servants in political fights, and making appointment and promotion decisions
in the civil service on the basis of political attitudes”).
52. These expansions came after Progressive Era reforms that, ultimately, put an end to the spoils
system and severely contracted party influence over appointments. See, e.g., Pendleton Civil Service
Reform Act, ch. 27, 22 Stat. 403 (1883) (codified as amended at 5 U.S.C. § 3304 (2012)).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
57
service merit system.53 During his term, as many as 1,128 positions were
filled through the Schedule C mechanism.54 During the Carter
administration, the Civil Service Reform Act of 1978 created the Senior
Executive Service (SES),55 a “corps of executives selected for their
leadership qualifications,”56 and provided that ten percent of total SES
positions could be filled with political appointees.57
Based on the expanded appointment freedom provided by Schedule C
and non-career SES positions, the number of positions filled outside of
civil service requirements (either at the discretion of the President or a
presidential designee, or upon nomination and approval by the Senate)
nearly doubled from 1960 to 1980, both in absolute terms and as a
percentage of non-military federal employees.58 Now, incoming presidents
are presented with between 3,000 and 4,000 political appointments that
must be filled across the Executive Branch.59
In addition, control over these personnel decisions has been centralized
over the past several decades. As Part II examines in more detail, there has
been an increase in the role of the White House in selecting and vetting
candidates for political positions, reducing the latitude given to cabinet
secretaries to select their staffs. Experiments with a more decentralized
personnel selection process—most notably during the Carter
administration—proved to generate unacceptable political costs. In the face
of those experiences, presidents have continually ratcheted up the
centralization of personnel decisions.
It is worth noting that agency structure, as well as norms and traditions,
can play a role in affecting the potential effectiveness of centralization and
politicization strategies. Politicization may be more difficult at the
“independent agencies” (such as the Securities and Exchange Commission)
in which for-cause removal and bipartisan appointment requirements limit
the ability of the President to place desired candidates in office. These
structural features are important and the subject of considerable legislative
53. PAUL CHARLES LIGHT, THICKENING GOVERNMENT: FEDERAL HIERARCHY AND THE
DIFFUSION OF ACCOUNTABILITY 45 (1995) (internal quotation marks omitted).
54. Id.
55. Civil Service Reform Act of 1978, Pub. L. No. 95-454, § 3131, 92 Stat. 1111, 1154–55.
56. U.S. OFFICE OF PERS. MGMT., GUIDE TO SENIOR EXECUTIVE SERVICE QUALIFICATIONS intro.
(2010), https://www.opm.gov/policy-data-oversight/senior-executive-service/reference-materials/guide
tosesquals_2010.pdf.
57. Civil Service Reform Act of 1978 § 3134.
58. David E. Lewis, The Contemporary Presidency: The Personnel Process in the Modern
Presidency, 42 PRESIDENTIAL STUD. Q. 577, 580 (2012).
59. Id. at 578.
2 LIVERMORE 45-134 (DO NOT DELETE)
58
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
bargaining,60 but they should not be overestimated. The President’s ability
to appoint the chair of independent commissions provides important
leverage,61 and once a majority of co-partisans is appointed, boards tend to
track presidential preferences.62 Conventions of independence also limit the
President’s ability to impose control over certain agencies, such as the
Federal Reserve Bank and the Food and Drug Administration.63 Further,
under the relevant executive orders, independent agencies have been
excused from the requirements of centralized executive review.64 So
although presidents may have broad incentives to pursue centralization and
politicization strategies across the executive, their effectiveness varies
across agencies depending on a variety of legal, institutional, and
bureaucratic factors.65
B. Presidential Control in Administrative Law
Executive restructuring to facilitate centralization and politicization has
largely taken place on the authority of the President, with occasional
support from Congress in the form of legislation. But doctrinal changes in
administrative law have also played a role in augmenting presidential
power over agencies. In the late 1970s and early 1980s, roughly
contemporaneously with the Reagan expansion of White House control
over agencies, several doctrinal developments scaled back judicial
oversight of agencies and facilitated more direct lines of centralized
accountability within the Executive Branch. These doctrinal developments
complemented efforts within the Executive Branch to reinforce presidential
control. The presidential-control orientation on the Court, however, was
never full-throated and it has faded in recent years,66 a point that will be
60. See generally DAVID E. LEWIS, PRESIDENTS AND THE POLITICS OF AGENCY DESIGN:
POLITICAL INSULATION IN THE UNITED STATES GOVERNMENT BUREAUCRACY 1946–1997 (2003).
61. See, e.g., RICHARD A. HARRIS & SIDNEY M. MILKIS, THE POLITICS OF REGULATORY
CHANGE: A TALE OF TWO AGENCIES (2d ed. 1996). See generally Kirti Datla & Richard L. Revesz,
Deconstructing Independent Agencies (and Executive Agencies), 98 CORNELL L. REV. 769 (2013).
62. Neal Devins & David E. Lewis, Not-So Independent Agencies: Party Polarization and the
Limits of Institutional Design, 88 B.U. L. REV. 459 (2008).
63. See Adrian Vermeule, Conventions of Agency Independence, 113 COLUM. L. REV. 1163,
1208 (2013).
64. Rachel E. Barkow, Insulating Agencies: Avoiding Capture Through Institutional Design, 89
TEX. L. REV. 15, 30–37 (2010). This exemption could be changed by a future president. See Datla &
Revesz, supra note 62.
65. If there is a zero-sum game between the President and Congress, then the independent
agencies might be less subject to presidential control, but more subject to congressional influence.
Justice Scalia argues that this is the case, at least for the Federal Communications Commission, in FCC
v. Fox Television Stations, Inc., 556 U.S. 502, 523 (2009).
66. Recently for example, in King v. Burwell, the majority opinion, authored by Chief Justice
Roberts and assented to by the liberal wing of the Court, stated that, on “question[s] of deep ‘economic
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
59
returned to in Part V. Given the extensive treatment of these issues
elsewhere, this introductory section provides only a brief survey that covers
a few of the relevant highlights.67
Under the “traditional model of administrative law,” agencies are
correctly understood as the “agents” of Congress, created and empowered
by statutes, which “provide[] rules that control and limit the agency’s
exercise of its authority”; when agencies fail to comply with these rules,
aggrieved parties can make recourse to independent courts, which ensure
conformity between agency action and statutory purposes.68 The exercise of
discretion is legitimated on the democratic foundation of statutory
authorization and the presumed separation of administrative decision
making from politics.69
As faith in impartial expertise waned and outsider groups pressed for
greater inclusion in administrative processes, a “reformation” period began
that was marked by expanded standing rules, more probing judicial review,
and increasing procedural requirements for agency rulemaking.70 The goal
of this reformation was to create, through administrative law, pluralistic
agency processes that would proxy for the broader democratic
accountability that was thought to be missing from bureaucratic decision
making.71
The reformation was short lived, lasting from the late 1960s to the mid1970s. A number of important cases in the late 1970s and early 1980s
diminished the scope of judicial review and explicitly favored presidential
and political significance,’” the presumption of Chevron deference to agency decisions may not hold.
135 S. Ct. 2480, 2489 (2015).
67. See, e.g., Lisa Schultz Bressman, Beyond Accountability: Arbitrariness and Legitimacy in the
Administrative State, 78 N.Y.U. L. Rev. 461, 485–91 (2003) (providing an intellectual history of the
support within the legal academy for a President-centered approach to administrative law); Sidney A.
Shapiro, Administrative Law after the Counter-Reformation: Restoring Faith in Pragmatic
Government, 48 U. KAN. L. REV. 689 (2000) (arguing that steps by the executive and courts to reform
administrative law in the 1980s had a distinctly anti-regulatory cast).
68. Keith Werhan, The Neoclassical Revival in Administrative Law, 44 ADMIN. L. REV. 567, 569
(1992). Under one version of this account agencies are viewed as “mere transmission belt[s]” that
translated legislative goals into administrative action through the application of expertise. Richard B.
Stewart, The Reformation of American Administrative Law, 88 HARV. L. REV. 1669, 1675–78 (1975).
69. See generally id. at 1678. There is extensive literature on the politics/administration divide,
which continues to have contemporary supporters.
70. Id. This loss of faith had multiple sources, including fear of agency capture by special
interest groups and concern that agencies were failing to actively carry out their mandates to promote
broad public interests. See Ernesto Dal Bó, Regulatory Capture: A Review, 22 OXFORD REV. ECON.
POL’Y 203 (2006) (review of capture literature); Livermore & Revesz, supra note 34; Thomas W.
Merrill, Capture Theory and the Courts: 1967-1983, 72 CHI.-KENT L. REV. 1039, 1043 (1997); Stewart,
supra note 68, at 1681–83 (1975).
71. Stewart, supra note 68, at 1670.
2 LIVERMORE 45-134 (DO NOT DELETE)
60
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
power.72 In Vermont Yankee Nuclear Power Corp. v. Natural Resources
Defense Council, Inc.,73 decided in 1978, the Supreme Court held that
courts may not “impose upon [an] agency its own notion of which
procedures are ‘best’ or most likely to further some vague, undefined
public good,” but must instead stick to the requirements found in the text of
the Administrative Procedure Act.74
In Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc.75
in 1984, the Supreme Court announced a standard of deferential review for
agency interpretations of ambiguous statutes. In Chevron, the Court
specifically acknowledged the accountability of agencies to the President in
justifying this deferential standard of review.76 Decisions such as Buckley v.
Valeo in 1976 (which strictly limit any role for Congress in the
appointment of Executive Branch officials77) facilitated more direct lines of
accountability within the executive. I.N.S. v. Chadha in 1982, which struck
down the legislative veto, removed an important tool for Congress to exert
control over agencies.78 The Court’s standing decision in Lujan v.
Defenders of Wildlife invoked separation-of-powers concerns to limit
judicial interference in executive decision making.79 Several D.C. Circuit
Court cases in the same time period further protected the White House’s
ability to play a substantial role in influencing agency policy, mainly by
rejecting challenges to agency action on the basis that the influence of the
President or White House offices biased agency decision making.80
The Court’s transfer of power to the executive has, however, always
been tempered. Most important, the Court’s State Farm jurisprudence
allows probing analysis under the Administrative Procedure Act’s
“arbitrary and capricious” standard, and agencies face a quasi-procedure
72. Peter L. Strauss, Presidential Rulemaking, 72 CHI.-KENT. L. REV. 965 (1997). Richard Pierce,
Political Accountability and Delegated Power: A Response to Professor Lowi, 36 AM. U. L. REV. 391
(1987).
73. 435 U.S. 519 (1978).
74. Id. at 549. But see Richard B. Stewart, Vermont Yankee and the Evolution of Administrative
Procedure, 91 HARV. L. REV. 1805 (1978) (critiquing the decision). The courts still examine the record
for substantive adequacy under the APA’s “arbitrary or capricious” review.
75. 467 U.S. 837 (1984).
76. Id. at 865–66. Interestingly, the Court’s “democratic legitimacy” argument went beyond the
expertise-driven justifications that were forwarded in the Government’s briefs in that case. William N.
Eskridge, Jr. & Lauren E. Baer, The Continuum of Deference: Supreme Court Treatment of Agency
Statutory Interpretations from Chevron to Hamdan, 96 GEO. L.J. 1083, 1087 (2008).
77. Buckley v. Valeo, 424 U.S. 1 (1976).
78. 462 U.S. 919 (1982).
79. 504 U.S. 555 (1992).
80. Sierra Club v. Costle, 657 F. 2d 298 (D.C. Cir. 1981); State of New York v. Reilly, 969 F.2d
1147, 1152 (D.C. Cir. 1992). For a criticism of these decisions, see Lisa Heinzerling, Classical
Administrative Law in the Era of Presidential Administration: Response to Daniel Farber and Anne
Joseph O’Connell, The Lost World of Administrative Law, 92 TEXAS L. REV. SEE ALSO 171 (2014).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
61
requirement to generate voluminous records in support of their decisions.81
Even within the Chevron regime, there have been many high-profile
instances of courts overruling agency decisions that were clearly linked to
presidential priorities.82 Despite strong language in Vermont Yankee
admonishing the D.C. Circuit for its procedural innovations, the Court has
largely eschewed playing an oversight role on the matter, and has given the
lower courts relatively free rein to maintain their prior habits.83 Decisions
like U.S. v. Mead Corp. and, more recently, King v. Burwell, have reduced
the circumstances in which Chevron deference is applicable.84 Justice
Thomas has gone substantially further, arguing that Chevron deference
essentially amounts to unconstitutional delegation of judicial and
legislative power to agencies.85 Even Justice Scalia, once Chevron’s most
vociferous supporter,86 has taken to lamenting that “too many important
decisions . . . are made nowadays by unelected agency officials . . . rather
than by the people’s representatives in Congress.”87 On structural matters,
even though the Court has remained fairly pro-President (for example,
recently expanding limitations on Congress’s power to place restrictions on
removal), discomfort with presidential power was evidence in the Court’s
split decision in a case over recess appointments.88
Overall, there is an official story of administrative law in which courts
facilitate presidential strategies of politicization and centralization and then
defer to agency decisions on the grounds that the (democratically
accountable) President exercises control over agencies. The Court, at times,
has seen that official story through. But there are also strong crosscurrents
81. Motor Vehicles Mfrs. Ass’n v. State Farm, 463 U.S. 29 (1983).
82. See, e.g., Massachusetts v. EPA, 549 U.S. 497 (2007) (overturning interpretation of the Clean
Air Act to exclude greenhouse gases from regulation, a position favored by the George W. Bush
administration); FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120 (2000) (overturning
Clinton-era rule subjecting tobacco to FDA oversight).
83. See generally Paul R. Verkuil, Judicial Review of Informal Rulemaking: Waiting for
Vermont Yankee II, 55 TUL. L. REV. 418 (1981); Richard J. Pierce, Jr., Waiting for Vermont Yankee II,
57 ADMIN. L. REV. 669 (2005); Jack Beermann & Gary Lawson, Reprocessing Vermont Yankee, 75
GEO. WASH. L. REV. 856 (2007).
84. United States v. Mead Corp., 533 U.S. 218 (2001) (limited Chevron deference to situations in
which it “appears that Congress delegated authority to the agency generally to make rules carrying the
force of law, and [] the agency interpretation claiming deference was promulgated in the exercise of that
authority”); see supra note 66 (discussing King v. Burwell).
85. Michigan v. EPA, 135 S. Ct. 2699, 2713 (2015) (Thomas, J., concurring).
86. See Mead Corp., 533 U.S. at 239 (strongly objecting to limitation of Chevron deference)
(Scalia, J., dissenting).
87. EPA v. Homer City Generation, 134 S. Ct. 1584 (2014) (Scalia, J., dissenting).
88. Free Enterprise Fund v. Public Co. Accounting Oversight Bd., 561 U.S. 477 (2010)
(removal); N.L.R.B. v. Noel Canning, 134 S. Ct. 2550 (2014) (recess appointments).
2 LIVERMORE 45-134 (DO NOT DELETE)
62
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
of skepticism about presidential control and willingness to intercede when
a sufficient number of Justices decide that it is justified.89
C. Decline and Rebirth of American Political Parties
In the late 1970s, at roughly the same time the Court began issuing its
pro-presidential oversight decisions, a new breed of political parties
emerged that transformed the structure of American political life. This
section provides a brief overview of that transformation.
Since shortly after the founding, party rivalry in one form or another
has been a dominant feature of American politics.90 Political parties can be
defined generally as “groups organized for the purpose of achieving and
exercising political power.”91 These groups are frequently thought of as
embracing three basic elements: the party in the electorate (voters who are
loyal to and identify with the party); the party as organization (party
officials, professional staff, donors, and volunteers); and the party in
government (officeholders and candidates).92 There is a long literature
within political science that focuses on the role that political parties play in
democratic systems.93 Elements of political parties that have received
considerable attention include the role of parties in structuring legislative
politics,94 the makeup and influence of local party organizations,95 the role
89. See, e.g., Jody Freeman & Adrian Vermeule, Massachusetts v. EPA: From Politics to
Expertise, 2007 SUP. CT. REV. 51 (arguing that Massachusetts v. EPA can be read as the Court taking
on a greater “expertise protection” function).
90. See generally COHEN ET AL., supra note 10.
91. ENCYCLOPÆDIA BRITANNICA, MICROPÆDIA VOLUME VIII 84 (15th ed. 1975). Nancy
Rosenblum provides a normative definition:
[A] party is a group organized to contest for public office; it is avowed in its partisanship
and operates not conspiratorially but in public view; it is not an ad hoc coalition or
arrangement for vote trading and compromise on a specific issue, but an institution formed
for ongoing political activity; and it can claim a substantial number of followers . . . .
NANCY L. ROSENBLUM, ON THE SIDE OF THE ANGELS: AN APPRECIATION OF PARTIES AND
PARTISANSHIP 20 (2008).
92. V.O. KEY JR., POLITICS, PARTIES, AND PRESSURE GROUPS (5th ed. 1964).
93. See, e.g., JOSEPH A. SCHUMPETER, CAPITALISM, SOCIALISM, AND DEMOCRACY (1942)
(explaining that parties facilitate competition by politicians for popular support); MARTIN SHEFTER,
POLITICAL PARTIES AND THE STATE: THE AMERICAN HISTORICAL EXPERIENCE (1994).
94. See GARY W. COX & MATHEW D. MCCUBBINS, LEGISLATIVE LEVIATHAN (2d ed. 2007);
GARY W. COX & MATHEW D. MCCUBBINS, SETTING THE AGENDA (2005). Work on “conditional party
government” covers parties in the legislature. See John H. Aldrich & David W. Rohde, The Republican
Revolution and the House Appropriations Committee, 62 J. POL. 1 (2000).
95. See GARY C. JACOBSON, THE POLITICS OF CONGRESSIONAL ELECTIONS 22–25 (8th ed.
2013); DAVID R. MAYHEW, CONGRESS: THE ELECTORAL CONNECTION (2d ed. 2004) [hereinafter
ELECTORAL CONNECTION] (explaining that candidates had to appeal to local constituents, regardless of
policy affiliation); DAVID R. MAYHEW, PLACING PARTIES IN AMERICAN POLITICS (1986) [hereinafter
PLACING PARTIES] (studying local party organizations and finding a long-term decline).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
63
of parties in presidential elections,96 and the importance of political parties
for structuring voters’ policy preferences.97
This literature teaches that the organization, function, and folkways of
political parties evolve in the face of cultural, technological, and legal
change. Political scientists sometimes refer to “party systems” that channel
political conflict through established party institutions over the course of
reasonably well-defined time periods. When a party system is no longer
adequate to the challenges of the day, it is swept away in a realignment that
fundamentally alters party composition as regional, ideological, and
interest group blocs shift allegiances.98 The election of William McKinley
in 1896 and FDR in 1932 are typically seen as such realignments.99
Traditional party organizations operated as the foundation for what Joel
H. Silbey characterizes as a “unique, partisan era” that lasted from the late
1830s into the 1890s.100 This form of party organization got its start when
new technologies and political alliances led to the creation of the “mass
mobilization” political party by Martin Van Buren and the other organizers
of the Democratic Party in the 1830s.101 The mass parties were the first to
engage everyday voters in party activities, and sought to actively cultivate
party identification among non-elites. They were organized primarily along
local lines, were not highly ideologically driven, and were strongly
motivated by the desire to generate patronage benefits for party loyalists—
most famously through the “spoils” system of delivering government jobs
in exchange for party services.102
96. See COHEN ET AL., supra note 10.
97. See DONALD GREEN, BRADLEY PALMQUIST & ERIC SCHICKLER, PARTISAN HEARTS AND
MINDS 204–29 (2002).
98. See generally Mark Tushnet, Foreword: The New Constitutional Order and the Chastening
of Constitutional Aspiration, 113 HARV. L. REV. 29 (1999). There is some resonance between partysystem transitions, realignment elections, and Ackerman’s periods of constitutional politics. See 1
BRUCE ACKERMAN, WE THE PEOPLE: FOUNDATIONS (1991); 2 BRUCE ACKERMAN, WE THE PEOPLE:
TRANSFORMATIONS (1998); 3 BRUCE ACKERMAN, WE THE PEOPLE: THE CIVIL RIGHTS REVOLUTION
(2014). At the very least, Ackerman’s constitutional moments match up with several of the party system
transitional periods. David R. Mayhew argues that the party system framework, organized around
realignment elections, is not a particularly informative way to analyze political development. See
DAVID R. MAYHEW, ELECTORAL REALIGNMENTS: A CRITIQUE OF AN AMERICAN GENRE 4–5 (2002).
This Article takes no particular position on this debate; the transition from traditional party
organizations to contemporary parties may not best be thought of as having occurred through
realignment elections. What is important is that the transition occurred sometime after the New Deal
was in full swing and before President Reagan’s election in 1980.
99. Ending the third and fourth party systems, respectively. See generally supra note 98.
100. JOEL H. SILBEY, THE AMERICAN POLITICAL NATION, 1838–1893, at 238 (1991) (describing
heyday of mass political parties and their effects on political life).
101. See generally JOEL H. SILBEY, MARTIN VAN BUREN AND THE EMERGENCE OF AMERICAN
POPULAR POLITICS (2002).
102. See SILBEY, supra note 100.
2 LIVERMORE 45-134 (DO NOT DELETE)
64
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
A variety of factors conspired to drive traditional party organizations
from the heart of American political life. During the Progressive Era, there
were sustained efforts to undermine prior forms of political association that
were thought to be corrupt and undemocratic.103 Progressive reforms, such
as the secret ballot and expansion of civil service requirements, reduced the
power of party bosses.104 Expansion of the regulatory state “took over by
policy and through entitlement the kinds of services local machines had
once made available.”105 After a long period of “mixed forms” that lasted
into the mid-twentieth century, whereby parties competed with alternative
political organizations including interest groups and administrative
agencies,106 traditional parties were “greatly weakened.”107
During the latter half of the twentieth century, legal, organizational,
and technological changes finally led to the death of traditional party
organizations and the birth of contemporary parties. John H. Aldrich gives
perhaps the definitive account.108 Intraparty differences, especially over the
issue of civil rights, which had been submerged by the economic crisis of
the Great Depression and the exigencies of World War II, began to
resurface and cause considerable tension within the parties. The final blow
came when traditional party organizations lost the last bastion of their
power, the “effective monopoly [they held] on the resources—the capital,
the labor, and the flow of information—that were necessary to run an
effective campaign.”109 According to Aldrich, the 1960 presidential
election ejected traditional party organizations from that stronghold when
John F. Kennedy created a personal campaign organization in his
presidential bid, rather than relying on the party organization. New
technologies, including the spread of television to many American
households and the availability of relatively cheap air travel, greatly
reduced the costs of directly communicating with voters and local elites. A
“new breed” of party activists, motivated not by patronage opportunities
but instead by demands for policy change, could be recruited to donate time
103. Richard Jensen, Party Coalitions and the Search for Modern Value: 1820–1970, in
EMERGING COALITIONS IN AMERICAN POLITICS (Seymour Martin Lipset ed., 1978).
104. See ALDRICH, supra note 6, at 281.
105. Id. Sidney Milkis has examined in detail how the rise of the administrative state has
transformed partisanship. See SIDNEY M. MILKIS, THE PRESIDENT AND THE PARTIES: THE
TRANSFORMATION OF THE AMERICAN PARTY SYSTEM SINCE THE NEW DEAL 4 (1993) (arguing that
political parties have acted, since the time of the founding, to thwart presidential administration and that
the New Deal administrative revolution was designed, in part, to “transcend[]” party politics).
106. SILBEY, supra note 100, at 241.
107. ALDRICH, supra note 6, at 167.
108. See id.
109. Id. at 282.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
65
and money to an individual candidate, rather than a party.110 Kennedy was
able to construct his own organization based on his personal charisma,
family connections, and policy vision, rather than having to rely on a
traditional party organization for electoral services.111
The decline continued after the 1960 election. After its turbulent 1968
convention, the Democratic Party drew up a set of internal reforms that
vastly increased the role of primaries in the selection of the presidential
nominee.112 Those reforms were followed by a similar set for the
Republicans. Campaign finance reform and alternative fundraising
channels reduced the financial role of traditional party organization.113 A
class of professional campaign operators, skilled in the new technologies of
mass communication, provided the services once dominated by parties.114
The electoral infrastructure of the parties was allowed to decay.115 In the
legislature, the rise of candidate-centered elections cut the cord between
members of Congress and their party, freeing members to establish a more
direct “electoral connection” with their constituents and reducing pressure
to toe the party line.116 There was a decline in partisan voting in Congress
and an increase in the activity of alternative voting blocs, such as the
“conservative coalition” of southern Democrats and conservative
Republicans that blocked (or attempted to block) liberal initiatives. In the
electorate, there was a rise of “independent” voters, and the value of party
as a “signal” to voters about the policy positions of affiliated politicians
was reduced.117 Observers of these collective trends argued that they
together sounded the death knell of American political parties.118
110. See KIRKPATRICK, supra note 2.
111. See ALDRICH, supra note 6, at 283.
112. HERBERT B. ASHER, PRESIDENTIAL ELECTIONS AND AMERICAN POLITICS 7 (rev. ed. 1980).
113. John H. Aldrich, Presidential Campaigns in Party- and Candidate-Centered Eras, in
UNDER THE WATCHFUL EYE 59, 69–71 (Mathew D. McCubbins ed., 1992).
114. See PLACING PARTIES, supra note 95, at 7. Even traditional outreach techniques (such as
door-to-door canvassing) that were traditional party organization specialties have become technologyand data-driven. See SASHA ISSENBERG, THE VICTORY LAB: THE SECRET SCIENCE OF WINNING
CAMPAIGNS (2012).
115. For example, after taking office, President Lyndon B. Johnson imposed severe cuts on the
Democratic National Committee. Sidney M. Milkis, The Presidency and Political Parties, in THE
PRESIDENCY AND THE POLITICAL SYSTEM 341, 348 (Michael Nelson ed., 8th ed. 2006).
116. See ELECTORAL CONNECTION, supra note 95.
117. ALDRICH, supra note 6, at 262–63.
118. See DAVID S. BRODER, THE PARTY’S OVER: THE FAILURE OF POLITICS IN AMERICA 251–54
(1971); WALTER DEAN BURNHAM, CRITICAL ELECTIONS AND THE MAINSPRINGS OF AMERICAN
POLITICS (1970); Jack Dennis, Trends in Public Support for the American Party System, 5 BRIT. J. FOR
POL. SCI. 187 (1975); Gerald M. Pomper, The Decline of the Party in American Elections, 92 POL. SCI.
Q. 21 (1977); see also Lawrence C. Dodd & Bruce I. Oppenheimer, The House Transition, in
CONGRESS RECONSIDERED 31, 50 (Lawrence C. Dodd & Bruce I. Oppenheimer eds., 2d ed. 1981)
(describing parties in Congress as “‘phantoms’ of scholarly imagination”).
2 LIVERMORE 45-134 (DO NOT DELETE)
66
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
Today, “[n]o one worries any more about the irrelevance of parties.”119
The decline of traditional party organizations was not the end of American
parties, but was instead the beginning of a transformation that led to the
rise, like “a phoenix . . . from the ashes,” of contemporary parties.120 This
new party form grew gradually over the course of the 1970s and 1980s, and
their nature remains contested in the political science literature (a point that
will be returned to in Part III.B). But there is general agreement about some
of the basic characteristics that describe the parties that currently dominate
the political scene. These characteristics affect how contemporary parties
manifest in the electorate, in government, and as organizations.121 They
include the following:
Contemporary parties are programmatic. This feature of parties
could be referred to as homogeneity, polarization, or coherence.
This Article adopts the word “programmatic” to describe the
phenomenon. In the electorate, voters now associate political
parties with policy programs: Democrats promote “liberal” policies
such as universal health care, affirmative action, and more stringent
pollution control; Republicans promote “conservative” policies
such as tax cuts, right-to-work laws, and deregulation. Voters who
affiliate with a party are now more likely to share the
programmatic positions that they associate with their party.122 In
the Legislature, the two parties have relatively clear policy
programs, and members largely adhere to those programs in their
voting.123 Organizationally, programmatic parties are closely
associated with policy-oriented activists and groups that provide
labor, money, and channels of communication with affiliated
voters.
Contemporary parties are nationally oriented. Traditional party
organizations were collections of relatively autonomous state and
local parties, with national party institutions that mediated and
119. ALDRICH, supra note 6, at 320.
120. Milkis, supra note 115, at 353.
121. KEY, supra note 92 (on tripartite structure of parties).
122. There are contested views about how the link between party programs and voter
identification came about. See, e.g., MATTHEW LEVENDUSKY, THE PARTISAN SORT: HOW LIBERALS
BECAME DEMOCRATS AND CONSERVATIVES BECAME REPUBLICANS (2009).
123. See ALDRICH, supra note 6, at 202–54; Richard Fleisher & John R. Bond, The Shrinking
Middle in the U.S. Congress, 34 BRIT. J. POL. SCI. 429 (2004); THE PEW RESEARCH CENTER, PARTISAN
POLARIZATION SURGES IN BUSH, OBAMA YEARS (2012), http://www.people-press.org/files/legacypdf/06-04-12%20Values%20Release.pdf.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
11/5/2015 9:20 AM
Political Parties and Presidential Oversight
67
coordinated but did not lead.124 Organizationally, that structure has
been reversed, and state and local party organizations now rely
heavily on the national committees, which have become
increasingly sophisticated, for financial, technological, and
logistical support. In the electorate, there is far less regional
variation among party-affiliated voters.125 In government, there has
been a shift in the relative importance of constituent services
versus the promotion of national policy. Even state and local
officials now use their offices to engage in debates over national
policy issues,126 presumably at the expense of more traditional
local concerns.
Contemporary parties are professionalized. Electoral campaigns,
and especially presidential campaigns, are now high-tech affairs
that rely on a dizzying array of experts in public relations,
marketing, social media, fundraising, logistics, polling, data
mining, and many other fields.127 To be effective electorally and
organizationally, parties now rely heavily on a cadre of campaign
professionals who are affiliated with an official party organization
or an associated group, or who operate as free agents within the
sphere of one or the other of the parties. In government, officials
rely on political appointees with expertise in law, social science,
information technology, national security, and communications
(inter alia) to navigate the complex policy environment of the
modern state.128
There is much more that can be (and has been) said about contemporary
political parties.129 Other important features of the partisan landscape
include: communications technologies, legislative rules, super donors
contributing to super PACs, constitutional protection for independent
expenditures, campaign finance reform, regional sorting according to
partisan affiliation, personality- and gaffe-driven campaigns, voter
124. KEY, supra note 92.
125. An important piece of this is the change in behavior of southern Democrats. Cf. ALDRICH,
supra note 6, at 241.
126. See generally Jessica Bulman-Pozen, Partisan Federalism, 127 HARV. L. REV. 1077 (2014).
127. ISSENBERG, supra note 114.
128. This point is discussed more thoroughly in Part II. It is not generally observed in the
political science literature, but follows similar logic.
129. See Joseph Fishkin & Heather K. Gerken, The Two Trends That Matter for Party Politics,
89 N.Y.U. L. REV. ONLINE 32 (2014), http://www.nyulawreview.org/sites/default/files/pdf/
nyulawreviewonline-89-32-fishkin-gerken.pdf (discussing the rise of “shadow parties” and the decline
of “formal parties”).
2 LIVERMORE 45-134 (DO NOT DELETE)
68
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
inattention, cable TV and social media Balkanization, the rise of Tea Party–
contested primaries.
But it is not necessary to have a complete survey of this landscape
before we proceed. The most important takeaway from this section is in the
broad strokes. Political parties now are different from those that structured
American politics for a century and a half and that still exerted
considerable influence when the modern administrative state was
constructed during the Progressive Era and New Deal and for several
decades thereafter.130 The patronage-oriented, locally autonomous, lowtech traditional party organizations of the past have disappeared. After a
period of lull, in their place appeared contemporary parties that are
programmatic, nationally oriented, and professionalized. As will be argued
below, this transformation has had important consequences for the role of
the President in the administrative state.
II. THE MECHANICS OF PRESIDENTIAL OVERSIGHT
This Part focuses on the descriptive question of how contemporary
parties help presidents effectuate control over the administrative state.
Although the centralization and politicization strategies discussed in Part
I.A have the potential to increase presidential control over administrative
agencies, they are by no means easy strategies to execute.131 Contemporary
parties have the potential to lower the costs of these strategies, which may
help account for the functional success of presidential administration over
the past several decades.
A. People Are Policy
The principal–agent model has frequently been applied to relations
between the President and civil servants at executive agencies.132 It can also
apply internally within the White House, and between the President and
130. A variety of scholars in history and political science have examined the dynamic
relationship between political parties and government institutions more generally during the early
periods of the administrative state. See, e.g., DANIEL P. CARPENTER, THE FORGING OF BUREAUCRATIC
AUTONOMY: REPUTATIONS, NETWORKS, AND POLICY INNOVATION IN EXECUTIVE AGENCIES, 1862–
1928 (2001); SCOTT C. JAMES, PRESIDENTS, PARTIES AND THE STATE (2000); WILLIAM J. NOVAK, THE
PEOPLE’S WELFARE: LAW AND REGULATION IN NINETEENTH-CENTURY AMERICA (1996); THEDA
SKOCPOL, PROTECTING SOLDIERS AND MOTHERS (1992); STEPHEN SKOWRONEK, BUILDING A NEW
AMERICAN STATE: THE EXPANSION OF NATIONAL ADMINISTRATIVE CAPACITIES 1877–1920 (1982).
131. Centralization “has been both a boon and a bane to presidents.” PFIFFNER, supra note 43, at
3.
132. See, e.g., Michael A. Livermore, Cost-Benefit Analysis and Agency Independence, 81 U.
CHI. L. REV. 609, 668 (2014).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
69
political appointees at agencies.133 The President has residual power over
the central staff, including the power of hiring and (typically) firing. At
agencies, some political appointees serve at the pleasure of the President,
while others enjoy tenure-protected positions, most notably the
commissioners at independent agencies. Regardless of the formal power to
remove unsatisfactory appointees, as a practical matter, it is impossible for
the President to be aware of most of the work that goes on in the Executive
Office of the President, much less at agencies, and effectively monitoring
and supervising that work is out of the question. The President is under
incredible cognitive demands, with a calendar packed full of public
appearances, fundraisers, and political glad-handing, leaving scarce time
for reading briefing papers or conducting supervisory meetings.134 Simply
satisfying the decisional demands of the highest priority domestic and
foreign matters requires nearly superhuman levels of mental endurance. For
the President, monitoring is extremely costly, creating the preconditions for
principal–agent problems.
So-called presidential control over agencies, then, is functionally
exercised by presidential designees with little presidential supervision.135
On many questions, the President will not have any opinion about the
correct course of action, and in any case will not be consulted. The
decisions of the political appointee or appointees charged with making
them will be final, in effect if not formally. For the small number of
decisions that require the President to act personally, it will largely be on
the advice of, and on the basis of information provided by, that same group
of appointees. If a policy or political question cannot be sorted out at a
lower level of the executive and demands presidential attention, there is a
good reason. It is likely controversial, complex, politically fraught, or
133. See generally George A. Krause, The Secular Decline in Presidential Domestic Policy
Making: An Organizational Perspective, 34 PRESIDENTIAL STUD. Q. 779 (2004) (attributing the decline
in presidents’ domestic policymaking to an increase in size—and therefore unwieldiness—of the
presidential bureaucracy).
134. Cynthia R. Farina, False Comfort and Impossible Promises: Uncertainty, Information
Overload, and the Unitary Executive, 12 U. PA. J. CONST. L. 357 (2010).
135. Kagan recognized that her argument concerning presidential control sometimes elided the
difference between the President and the White House, noting that “often when I refer to ‘the President’
in this Article, I am really speaking of a more nearly institutional actor — the President and his
immediate policy advisors in OMB and the White House.” Kagan, supra note 3, at 2338. Kagan even
promised a “future article” that would “delve more deeply into . . . the black box of the EOP [and]
examin[e] this office as itself an administrative agency and explor[e] the relationships (in part
influenced by legal rules) among the President and all the EOP’s constituent parts.” Id. at 2338 n.352.
Terry Moe is more sanguine, conceding that politicization carries “knowledge demands” that are “not
negligible,” but arguing that “the president will find politicization irresistible” in part because “[i]t
assumes no sophisticated institutional designs and little ability to predict the future.” Moe, supra note
12, at 245.
2 LIVERMORE 45-134 (DO NOT DELETE)
70
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
impactful. These questions do not present easy or obvious answers.136
Given the difficulty of the questions and limited time and cognitive
resources, the policy alternatives, risk analyses, background briefings, and
political advice given by senior personnel strongly structure the President’s
choice set. The opportunity to present those alternatives, analyses,
briefings, and advice translate directly into influence over the ultimate
policy outcomes.
It is therefore essential that the individuals selected to fill positions
within the administration share the President’s preferences if centralization
and politicization are to serve the goals of presidential oversight. As the
Reagan transition team was fond of saying, “[p]eople are policy.”137 If
appointees do not share the President’s preferences, the most likely
outcome is that the decision will be made by the appointee, contrary to
presidential preferences, because the President does not know that the
decision is being made and the appointee does not know the President’s
preferences. Given the vast scope of the administrative state, the sheer mass
of decisions to be made, and the limited communication between the
President and even extremely senior executive officials, mutual ignorance
is the norm, not the exception. This risk is especially acute when
presidential preferences differ from the programmatic positions of his or
her party because appointees are likely to assume that the President
endorses the party’s program unless differences are clearly broadcast. Even
if appointees know where the President is likely to stand on an issue, they
have tools to thwart the President’s goals, if they so desire. Such appointees
might attempt to channel decisions away from presidential loyalists, “slow
walk”138 directives from above, short-change analysis,139 fail to offer policy
alternatives, activate political constituencies, leak information to the media,
or even resign their positions and engage in external advocacy, all of which
would make it difficult, if not impossible, for the President’s preferences to
be vindicated.
In addition to the problem of loyalty, there is a second problem of
competence. Even well-intentioned central staffers who believe that they
are carrying out the President’s policy agenda can make extremely poor
and politically damaging decisions. The Iran-Contra operation, run out of
the national security offices within the Executive Office of the President,
provides the classic example of centralized authority gone wrong. The
136. See Livermore & Revesz, supra note 34, at 1349.
137. PFIFFNER, supra note 14, at 94 (internal quotation marks omitted).
138. See Darren Samuelsohn, Brass Slow-Walk Sex-Assault Orders, POLITICO (Nov. 17, 2013,
4:15 PM), http://www.politico.com/story/2013/11/military-sex-assault-obama-099962.
139. See Jennifer Nou, Agency Self-Insulation Under Presidential Review, 126 HARV. L. REV.
1755, 1771 (2013).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
71
scheme, which involved selling arms to the Iranian army and using the
proceeds to fund operations by a paramilitary group in Nicaragua, was a
failure along every possible dimension and took an extraordinary political
toll on the Reagan administration.140 When political appointees are
incompetent, it creates a double problem for the President, because the
layer of removal provided by agency independence is not available. The
President gains no advantage from centralization and politicization if it
results in a substantial reduction of the quality of decision making on
important national security or domestic policy matters. In the complex
political, economic, and geopolitical environment in which the U.S.
government operates, presidents cannot afford to make substantial
sacrifices in competence for the sake of increased centralization of
executive decision making.
Balancing the requirements of loyalty and competence in making
personnel decisions is one of the most difficult challenges that presidents
face.141 The President comes into office with 3,000 to 4,000 positions to
fill,142 and the turnover of political appointees is quite high, requiring
constant hiring.143 A quarter of those positions require Senate confirmation,
opening the President to ridicule or scandal if nominees fail any number of
modern litmus tests.144 Many of the positions involve leadership and
personnel management, policy discretion, and public appearances, creating
a host of opportunities to undermine the administration’s policy
preferences, botch important government functions, and generally create
headaches for senior officials and embarrass the President. The effective
use of the power of politicization requires the identification of a very large
number of officials across a range of disciplines and professions who are
willing to uproot their lives for short-lived, stressful, high-workload
positions and who will implement the President’s policy agenda with little
to no supervision. The following section discusses the constraints and
incentives that structure how presidents attempt to carry out that task.
140. Larry Martz et al., “It Was My Idea”: Dropping His Didn’t Know Defense, Reagan Takes
Credit for Contra Aid, NEWSWEEK, May 25, 1987, at 16–19.
141. There also may be a tradeoff between centralization and competence if issue-specific
agencies tend to have greater expertise. See Barkow, supra note 64, at 34 (explaining that “[t]he
relationship between expertise and OIRA is . . . a complicated one”).
142. Lewis, supra note 58, at 578.
143. B. Dan Wood & Miner P. Marchbanks III, What Determines How Long Political
Appointees Serve?, 18 J. PUB. ADMIN. RES. & THEORY 375 (2007) (finding political appointees served
an average of 3.48 years, with a substantial group of shorter tenures, including 25% of appointees
serving less than one year, 46% serving less than two years, and only one third remaining by the end of
their third year).
144. See generally INNOCENT UNTIL NOMINATED: THE BREAKDOWN OF THE PRESIDENTIAL
APPOINTMENTS PROCESS 161 (G. Calvin Mackenzie ed., 2001); O’Connell, supra note 26
(documenting increase in time to confirmation and failure rates in recent presidencies).
2 LIVERMORE 45-134 (DO NOT DELETE)
72
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
B. Parties and Appointments
As the number of political appointees has expanded, presidents have
sought to centralize control over their selections. Whereas President
Eisenhower created the Schedule C mechanism to give cabinet-level
secretaries greater freedom over their personnel decisions,145 subsequent
administrations have exerted progressively greater centralized control.
President Kennedy’s three-person personnel staff has bloomed into the
contemporary presidential transition team staffed by hundreds of
professionals that attempts to subject every political appointee of any
consequence to centralized scrutiny.146 Beginning at least in the
administration of President Nixon, loyalty to the President and consistency
with the administration’s ideological direction have been given especially
high priority.147 The desire to fill non-career positions with individuals who
will implement the President’s vision is no secret; for example, the Office
of Personnel Management has stated quite directly that “[t]he President and
Presidential appointees expect Schedule Cs to represent the
administration’s goals, viewpoints and philosophies as their own.”148
As traditional party organizations declined, they exerted ever less
pressure on the White House in the area of personnel. During the years
when the spoils system dominated, parties played the central role in
allocating government employment opportunities, “parcel[ing] out control
over appointments to different factions within the party after the election,
with factions often given explicit control over a specific subset of
nominations either by agency or region.”149 Dominant players within the
parties—including members of Congress and high-ranking state or even
municipal officials, representatives of dominant interests (such as labor
unions or industrialists), and individuals holding official party positions—
used government positions as a way to reward loyal party service, pay back
campaign favors, and generally promote the electoral prospects of their
party as well as their personal or factional interests. This relationship lasted
through the Eisenhower administration.150
145. LIGHT, supra note 53, at 45.
146. Lewis, supra note 58, at 585.
147. See Donald P. Moynihan & Alasdair S. Roberts, The Triumph of Loyalty Over Competence:
The Bush Administration and the Exhaustion of the Politicized Presidency, 70 PUB. ADMIN. REV. 572,
573 (2010) (discussing personnel policies during the George W. Bush administration); Rudalevige &
Lewis, supra note 51, at 8.
148. OFFICE OF PERS. MGMT., SCHEDULE C QUESTIONS & ANSWERS (Q’S & A’S) 4 (July 7,
2003), http://www.dm.usda.gov/employ/exec/docs/schedule-c-qa.pdf [https://web.archive.org/web/
20130312145532/http://www.dm.usda.gov/employ/exec/docs/schedule-c-qa.pdf].
149. Lewis, supra note 58, at 584.
150. Id. at 585.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
73
The symbiotic relationship between electoral support and patronage
employment broke down in the mid-twentieth century alongside the
traditional party organizations that had facilitated that transaction.
Interactions between the White House and the in-power party transitioned
accordingly. Although prior administrations had worked closely with the
traditional party structures to manage the flow of patronage, the Kennedy
administration broke with past practice by establishing a separate
organization to manage staffing during the transition and by conducting
separate searches to fill patronage and policy positions.151 Subsequent
administrations have followed a similar course, with traditional party
structures exhibiting ever diminishing influence.152
But the decline of traditional party organizations has not resulted in a
free hand for the President over personnel. Instead, the “new constellation
of organizations that now dominate electoral politics . . . shap[e] and
constrain[] the Presidential Personnel Office in much the same way that
party organizations once did.”153 The demands from this constellation fall
into roughly three categories: diversity/representation, pure patronage, and
policy patronage. In his study of presidential appointments in the postMcGovern–Fraiser reform period, Thomas J. Weko focuses on the new
demands for diversity that confronted presidents. In the Carter
administration, the Presidential Personnel Office “was deeply engaged in
monitoring [appointments] to ensure that politically appropriate numbers of
women, African-Americans, and Hispanics were appointed.”154
Dissatisfaction over early appointments led to the Reagan personnel office
hiring “a set of . . . campaign aides to assist . . . in handling relations with
women’s groups [and] ethnic minorities (especially Hispanics).”155 In the
Clinton administration, the personnel team was “finely calibrated” with a
group of “liaisons to feminist, African-American, Hispanic, gay, and
disability groups”156 to collect resumes and ensure serious consideration of
identified candidates.
The consequences of this particular category of demands for
presidential control are unclear. Given the limited number of appointments
compared to the potential pool and the relative desirability of the positions,
it is not obvious that meeting diversity goals is a major constraint. These
151. See THOMAS J. WEKO, THE POLITICIZING PRESIDENCY: THE WHITE HOUSE PERSONNEL
OFFICE, 1948–1994, at 26 (1995).
152. Emulation of the Kennedy structure included creating separate processes for patronage and
expertise positions. See id.
153. Id. at 78.
154. Id. at 86–87.
155. Id. at 98.
156. Id. at 100.
2 LIVERMORE 45-134 (DO NOT DELETE)
74
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
goals may create tension within an administration over individual
appointments,157 however, and the personnel who implement diversity
priorities may have (or be perceived as having) loyalties that are divided
between the President and “causes and constituencies outside of the White
House.”158
A second class of demands is for what can be called pure patronage.
While the wholesale patronage that existed during the spoils system is no
more, retail patronage remains a task for presidential appointments. Many
of these demands come from applicants who “worked on the [President’s]
campaign, for a state party, a member of Congress, or [an] interest
group.”159 These patronage seekers are less interested in a paycheck and the
other basic perquisites of a position (although of course these have value)
than they are in “a job that will give them a rewarding work experience and
advance their career prospects, particularly within the party or its
constellation of related groups.”160 With the decline of traditional party
organizations, these demands now come to the White House, rather than
being filtered through external channels.
Demands for pure patronage create a substantial potential tension
between loyalty and competence. Professor David E. Lewis characterizes
patronage seekers as largely “young, politically ambitious, [with] limited
experience.”161 Another important class of potential patronage seekers is
campaign donors, especially the so called “bundlers” that put together large
pools of donations. The Center for Public Integrity found that 184 out of
556 bundlers that accounted for more than $50,000 in campaign
contributions to the 2008 Obama campaign, or their spouses, landed
positions in the administration.162 Of bundlers who raised at least $500,000,
nearly 80% were appointed to “key administration posts.”163 This group is
different from that of the young and inexperienced campaign worker, and
includes business owners, executives, and professionals.164 Nevertheless,
drawing from a highly limited pool of donors for important administration
157. Id. at 87 (general counsel at Department of Health, Education, and Welfare under Jimmy
Carter discussing such tension).
158. Id. at 102.
159. DAVID E. LEWIS, THE POLITICS OF PRESIDENTIAL APPOINTMENTS: POLITICAL CONTROL
AND BUREAUCRATIC PERFORMANCE 63 (2008); David E. Lewis, Revisiting the Administrative
Presidency: Policy, Patronage, and Agency Competence, 39 PRESIDENTIAL STUD. Q. 60 (2009).
160. Id.
161. Id.
162. Fred Schulte, John Aloysius Farrell & Jeremy Borden, Obama Rewards Big Bundlers with
Jobs, Commissions, Stimulus Money, Government Contracts, and More, CENTER FOR PUB. INTEGRITY
(June 15, 2011, 2:00 AM), http://www.publicintegrity.org/2011/06/15/4880/obama-rewards-bigbundlers-jobs-commissions-stimulus-money-government-contracts-and.
163. Id. (internal quotation marks omitted).
164. See generally id.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
75
positions creates substantial risks. These are perhaps best embodied by
Michael D. Brown, who was the head of the Federal Emergency
Management Agency (FEMA) during Hurricane Katrina. Brown’s work
history prior to his appointment at FEMA, which included a failed run for
Congress, local political positions in his home state of Oklahoma, and a
nine-year stint as the Commissioner of the International Arabian Horse
Association, came under sustained criticism in the wake of FEMA’s
perceived mismanagement of the disaster response.165
Perhaps ironically, the competence risks associated with pure
patronage may be greatest for agencies and issues that are most valued by
powerful groups within the party coalition. In an extensive analysis of
political appointees during recent administrations, Lewis finds that
patronage appointees tended to be targeted at agencies that are anticipated
to be in agreement with the current administration.166 For example,
conservative agencies, such as the Small Business Administration, were
targeted for patronage appointments during Republican administrations.167
So even though the smooth functioning of agencies that are aligned with
the interests of groups in the party coalition would seem to be one of the
benefits of gaining power, presidents’ patronage incentives appear to
interfere with this goal.
The final class of demands on presidential appointments is policy
patronage. Unlike pure patronage, which responds to individual desires to
burnish an ambitious young person’s resume or add an ambassadorship to
an already accomplished executive’s life experiences, policy patronage
responds to the demands of organized interests for appointees who can be
trusted to forward their programmatic goals.168 Policy patronage demands
may reflect broadly agreed-upon party priorities, but may also generate
intraparty disputes. For example, Weko describes conflict during the early
years of the Reagan administration over claims by “‘new right’ politicians,
publicists, and think tanks” that were concerned that “‘movement
conservatives’ were being given short shrift in favor of ‘Nixon-Ford
retreads.’”169 The head of Reagan’s personnel office was criticized for
failing to show “deep commitment to new right causes” and favoring
“‘proven managers’ drawn from the traditional Republican business
165. See Spencer S. Hsu & Susan B. Glasser, FEMA Director Singled Out by Response Critics,
WASH. POST, Sept. 6, 2005, at A1; see also Moynihan & Roberts, supra note 147, at 575.
166. See LEWIS, supra note 159, at 133 (using Schedule C appointments as proxy for
“patronage” and PAS as proxy for “policy”).
167. Id. at 133–36.
168. Cf. PAUL P. VAN RIPER, HISTORY OF THE UNITED STATES CIVIL SERVICE, 327 (1958)
(examining how expansion of administrative state during the New Deal created “a sort of intellectual
and ideological patronage rather than the more partisan type”).
169. WEKO, supra note 151, at 96 (internal quotation marks omitted).
2 LIVERMORE 45-134 (DO NOT DELETE)
76
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
constituency, rather than men and women who ‘could supply conservative
intellectual leadership to the bureaucracy.’”170 Especially after highly
contested primaries (such as the 2008 Obama/Clinton primary), policy
patronage demands may tilt toward intraparty power-allocation. But after
all elections, groups within the party coalition (including those with
conflicting interests) will jockey for appointees that they trust to forward
their interests, and these competing demands are negotiated and balanced
by the President’s personnel team.
It is in this last category of patronage, oriented toward satisfying the
policy demands of interest groups, where the transition to contemporary
parties has the most importance. Traditional party organizations were
experts at handling demands for diversity/representativeness and patronage.
Managing regional and ethnic claims to plum positions was the bread and
butter of traditional party organization and the payoff for hard-fought
electoral victories. Transferring those tasks to the presidency increases
cognitive loads in the White House and the potential for misfiring, but does
not seem to otherwise have a major effect on presidential administration.
As will be discussed in the next section, however, the transition to
contemporary parties, with their national, programmatic, and
professionalized focus, both creates increased demands for policy
patronage and provides tools for the President to settle those demands. If
used wisely, these tools have the potential to enhance presidential control
over agencies.
C. Performance, Politics, and the New Patronage
As described above, the primary presidential strategies for asserting
control over agencies hinge on the ability to identify competent and loyal
personnel to fill key policymaking positions across the federal
government.171 Incompetent or disloyal personnel do a presidential
administration no good, and in fact can cause serious problems.
Responding to this challenge is a central ingredient of a successful
presidency. Contemporary political parties provide presidents with four
important advantages in doing so: lack of a competing power center,
signaling value, policy expertise, and constituency balancing.
The first advantage is, in essence, a negative benefit—the lack of an
alternative power center on personnel decisions that competes with the
President for control over the allocation of political appointments. As
discussed in the previous section, traditional party organization once played
170. Id. at 97.
171. See supra Part II.A.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
77
a major role in personnel decisions made by the White House, with new
presidents essentially delegating many of these decisions to party
organization and party processes. With the decline of traditional party
organization, this competing center of power is no longer effective.
When a President enters office, a variety of demands come into the
White House, and they must be settled. In particular, there may be
differences of opinion on policy direction within the network of affiliated
interest groups that make up contemporary parties. Personnel decisions are
a way to accommodate conflicting demands or decisively settle disputes
and strike out on a definitive path. Absent some other strong coordinating
mechanism, it falls to the President and presidential loyalists in the most
senior political positions (such as the Chief of Staff and head of the
personnel office) to make the final call. These decisions are made more or
less free of any concerns about what the chair of the national party
committee might think, resulting in greater leeway for the President and his
or her agents.
Second, the programmatic nature of contemporary political parties
provides valuable signals about the views of potential appointees.
Presidents can better predict the behavior of (and rely on) their appointees
if potential appointees are loyal to the party program across a broad policy
domain. Of course, contemporary political parties have not eliminated
intraparty disputes, which remain heated. But by many measures, the
overlap between the parties—both within government and within the
electorate—has largely disappeared. When selecting an appointee from a
pool of applicants who have built careers with interest groups or politicians
affiliated with the Republican Party, there is little concern that an
administration will inadvertently choose a candidate to the left of a
similarly situated candidate whose experience was with Democraticaffiliated organizations or individuals. Because parties have adopted
opposed positions on almost every issue of broad public concern, and
consistency with those positions has become a hallmark of party
identification, the signaling value of party affiliation is large.
Professors Devins and Lewis examine the role of party affiliation in
determining the behavior of bipartisan commissions on independent
agencies.172 They find that the party of the commissioner is a much stronger
predictor of behavior than the party of the appointing President; when
bipartisan requirements require that a Republican President nominate a
Democratic commissioner, the commissioner does not act as a “turncoat”
but votes in line with Democratic priorities.173 The result is that presidents
172. See Devins & Lewis, supra note 62.
173. Id. at 461.
2 LIVERMORE 45-134 (DO NOT DELETE)
78
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
cannot gain control over bipartisan commissions until they have been in
office long enough to nominate a majority from the same party, but once
that occurs, “today’s independent agencies are more likely to agree with
presidential preferences” because of party polarization.174
The third advantage offered by contemporary political parties is policy
expertise. Political scientists have long discussed the importance of “issue
networks” in contemporary governance.175 These issue networks include
the think tanks, advocacy organizations, foundations, and private lobbying
operations that now dominate the Washington, D.C., landscape.176
Although some of these actors are putatively nonpartisan, many of them
nonetheless have reliable party affiliations; for example, the Center for
American Progress is a Democratic-affiliated entity, while the Cato
Institute is affiliated with the Republican Party. These issue networks
cultivate a group of “policy activists” who are “intellectual[ly] or
emotional[ly] commit[ted]” to programmatic goals and who gain deep
agency or area-specific technical expertise.177
This intellectual infrastructure can provide a variety of expertiserelated services. Presidents can rely on issue networks to identify
appointees with the depth of domain-specific knowledge and prior
exposure to relevant policy questions needed to quickly translate their new
authority into programmatic policy output. In addition, issue networks
develop recommendations for friendly administrations and political
appointees can rely on their connections for feedback and information.
Further, when a hostile administration is in office, these organizations
generate criticism that can be translated into a programmatic vision during
electoral campaigns.
Issue networks are also involved in a fourth advantage offered by
contemporary political parties: they help balance potentially conflicting
constituencies. Parties are coalitions that, by necessity, include substantial
diversity. As a consequence, there are always latent fracture lines.
174. Id.; see also Daniel E. Ho, Congressional Agency Control: The Impact of Statutory Partisan
Requirements on Regulation (Feb. 12, 2007) (unpublished manuscript), http://dho.stanford.edu/
research/partisan.pdf (finding that cross-party appointees are more responsive to their party than to the
President who appointed them). See generally Pablo T. Spiller & Santiago Urbiztondo, Political
Appointees vs. Career Civil Servants: A Multiple Principals Theory of Political Bureaucracies, 10 EUR.
J. POL. ECON. 465 (1994) (examining the role of the party in affecting principal–agent relationships
among career and political appointees at agencies, the President, and Congress).
175. See Hugh Heclo, Issue Networks and the Executive Establishment, in The NEW AMERICAN
POLITICAL SYSTEM 87 (Anthony King ed., 1978). See generally James E. Skok, Policy Issue Networks
and the Public Policy Cycle: A Structural-Functional Framework for Public Administration, 55 PUB.
ADMIN. REV. 325 (1995) (arguing that issue networks play a fundamental role in determining the
agenda for political institutions).
176. See generally THOMAS MEDVETZ, THINK TANKS IN AMERICA (2012).
177. Heclo, supra note 175, at 102.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
79
Personnel decisions that inexpertly weigh competing concerns within the
party can lead to substantial challenges for a president.178 In the
contemporary Democratic Party, unions and environmentalists are two
important constituencies with interests that conflict when environmental
protections threaten incumbent workers in unionized industries. On the
Republican side, issues such as gay rights pit libertarians against social
conservatives, two important constituencies that define the contemporary
party. For both parties, there are issue-network entities that help mediate
potential conflicts. On the Democratic side, an example is the BlueGreen
Alliance. This organization was established by leading labor and
environmental organizations specifically to help facilitate discourse and
compromise between those two groups.179 On the Republican side, there
are entities such as the American Conservative Union, which sponsors the
annual Conservative Political Action Conference—a forum where areas of
policy disagreement are aired and discussed, with a broader understanding
that maintaining the party coalition is of primary importance.180
These coalition-maintaining efforts help presidents avoid intraparty
conflicts over personnel decisions and cultivate potential appointees who
have expertise in balancing conflicting demands within the party. Through
experience with the BlueGreen Alliance and similar activities, policy
activists within the environmental movement develop personal
relationships within organized labor and form policy perspectives that
178. For example, Republican EPA Administrators have traditionally been called on to balance
the desires of the northeastern wing of the party, which had strong environmentalist tendencies, against
the antiregulatory demands of industry and libertarian western elements of the party. This intraparty
conflict often results in intra-administration conflict. After Reagan EPA Administrator Ann Gorsuch—
who took a hostile approach to the agency’s mission—was forced to resign amid scandal, the
administration faced sufficient pressure on environmental issues that it ultimately appointed William D.
Ruckelshaus, an environmentalist icon from the Nixon administration. The EPA Administrator under
George H.W. Bush, William K. Reilly, regularly clashed with other political appointees in the White
House, especially the Council on Competitiveness, chaired by Vice President Dan Quayle. See William
K. Reilly: Oral History Interview, U.S. ENVTL. PROTECTION AGENCY (Sept. 1995),
http://www2.epa.gov/aboutepa/william-k-reilly-oral-history-interview. Christine Todd Whitman, EPA
Administrator under George W. Bush, also clashed with the White House, and eventually resigned after
being overruled on a major environmental rulemaking. See Jo Becker & Barton Gellman, Leaving No
Tracks, WASH. POST, June 27, 2007, at A1.
179. See generally About Us, BLUEGREEN ALLIANCE, http://www.bluegreenalliance.org/about
(last visited Sept. 25, 2015) (“The BlueGreen Alliance unites America’s largest labor unions and its
most influential environmental organizations to identify ways today’s environmental challenges can
create and maintain quality jobs and build a stronger, fairer economy.” (emphasis omitted)).
180. For example, conflicts within the Republican coalition over the issue of gay rights
periodically break out in the context of the American Conservative Union’s annual conference, with gay
rights–oriented groups arguing that they have been excluded from this premier party event. See, e.g.,
David McCabe, Gay GOP Group Says It’s Being Excluded from CPAC, THE HILL (Feb. 20, 2015, 2:39
PM), http://thehill.com/blogs/blog-briefing-room/233356-gay-gop-group-says-theyre-being-excludedfrom-conservative-confab. These conflicts can be understood as a means of mediating the inherent
tensions within a party that includes libertarians as well as social conservatives.
2 LIVERMORE 45-134 (DO NOT DELETE)
80
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
incorporate these sometimes competing demands. Similarly, exposure to
the wide range of Republican Party constituencies through entities such as
the American Conservative Union helps cultivate foreign policy hawks
who understand the concerns of deficit hawks and tax hawks—all of whom
have conflicting policy desires but must learn to collaborate to develop a
party program that is mutually acceptable.
Contemporary parties that can supply a group of readily identifiable
technocrats who are committed to a consistent set of policy initiatives
associated with a party program and who have experience balancing the
interests of the party’s various constituencies provide an extremely
valuable service for presidents as they exercise their appointment power. It
is hard to image anything more useful for an incoming Executive than an
extensive network of committed, well-trained, enthusiastic, and loyal
potential personnel to fill vital, difficult-to-supervise, and controversial
roles. To the degree that contemporary parties are well suited to supplying
that network, they will facilitate, to a great degree, the reality of
presidential oversight.
III. PARTIES AND PRESIDENTIAL REPRESENTATION
Critics of the trends toward presidential administration argue that an
expansive role for the President, political appointees, and the White House
threatens administrative, constitutional, and democratic values.181
Defenders of presidential power frequently respond to these critiques by
pointing to the virtues of the presidency, and in particular to the President’s
unique connection to the national electorate. Under this presidential
representation hypothesis, greater presidential influence in the
administrative state promotes majoritarian values by subjecting agencies—
at least indirectly—to electoral accountability.182 In Chevron, the Court
adopted a version of this logic, arguing that agencies may “properly rely
upon the incumbent administration’s views of wise policy to inform its
judgments” because, “[w]hile agencies are not directly accountable to the
people, the Chief Executive is, and it is entirely appropriate for this
181. See DAVID SCHOENBROD, POWER WITHOUT RESPONSIBILITY: HOW CONGRESS ABUSES THE
PEOPLE THROUGH DELEGATION (1993); Barkow, supra note 64, at 23; Lisa Heinzerling, Statutory
Interpretation in the Era of OIRA, 33 FORDHAM URB. L.J. 1097 (2006).
182. See generally Bressman, supra note 67 (discussing the majoritarian defense of presidential
control); DeMuth & Ginsburg, supra note 49; Kagan, supra note 3; see also Daniel B. Rodriguez,
Management, Control, and the Dilemmas of Presidential Leadership in the Modern Administrative
State, 43 DUKE L.J. 1180, 1193–94 (1994) (arguing that the national constituency provides the
President with a comparative advantage for regulatory reform because the President is “less vulnerable
to targeted appeals by interest groups” and because the President is “[n]ot plagued by the difficulties
associated with collective decisionmaking”).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
81
political branch of the government to make such policy choices.”183 Similar
notes remain frequently sounded by the courts.
But the normative attractiveness of this justification for presidential
power depends on the structure of presidential electoral politics, including
the role of contemporary parties in nominating and supporting presidential
candidates. If the nomination process screens for candidates with
preferences that are substantially different from those of typical voters, or if
the electoral process favors narrow, organized interests at the expense of
the public, the presidential representation rationale for presidential
oversight of agencies loses much of its normative attractiveness. As strong
parties reemerged at the end of the twentieth century, a debate arose in the
political science literature about whether contemporary parties are
dominated by politicians or groups. Research on that question bears
directly on the viability of the presidential representation hypothesis, and is
the subject of this Part.
A. A Simple Model of Presidential Representation
The presidential representation hypothesis has been highly influential,
taking on the aspect of a “mystique” or “unifying theory” in defense of
expanded presidential power over agencies.184 In addition to the Court in
Chevron, the presidential representation basis for presidential power over
agencies has been embraced in some form by a long list of luminaries
across the political spectrum, including Justice Elena Kagan,185 Judge
Frank Easterbrook,186 Judge Douglas Ginsburg,187 and Professors Jerry L.
Mashaw,188 Lawrence Lessig, and Cass Sunstein.189
The normative force of the claim stems from an understanding of
presidential control as a means of binding agency decisions to majority
183. Chevron, U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 865 (1984).
184. See Bressman, supra note 67, at 478 (“unifying theory”); Thomas O. Sargentich, The
Emphasis on the Presidency in U.S. Public Law: An Essay Critiquing Presidential Administration, 59
ADMIN. L. REV. 1, 35–36 (2007) (referring to a “presidential mystique” that supports expanded
presidential control over agencies); see also HAROLD SEIDMAN, POLITICS, POSITION, AND POWER: THE
DYNAMICS OF FEDERAL ORGANIZATION (2 ed. 1975).
185. Kagan, supra note 3.
186. Frank H. Easterbook, The State of Madison’s Vision of the State: A Public Choice
Perspective, 107 HARV. L. REV. 1328, 1341 (1994).
187. DeMuth & Ginsburg, supra note 49.
188. Jerry L. Mashaw, Prodelegation: Why Administrators Should Make Political Decisions, 1
J.L. ECON. & ORG. 81, 90 (1985).
189. Lawrence Lessig & Cass R. Sunstein, The President and the Administration, 94 COLUM. L.
REV. 1, 101 (1994).
2 LIVERMORE 45-134 (DO NOT DELETE)
82
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
preferences.190 In a highly simplified way, the goal of political oversight
could be thought of as translating the preferences of the median voter into
regulatory policy.191 In a basic model of politics, if a slate of regulatory
proposals were offered up for general majority vote, the proposal closest to
the median voter’s preferences would dominate.192 If the President’s
preferences match those of the median voter, then making agencies
responsive to presidential preferences will replicate the outcomes that
would have been generated by a plebiscite.193
Of course, the President as a proxy for the median voter is a somewhat
crude expression of presidential representation hypothesis. Presidents have
access to information that the median voter does not, and their role is not
merely to serve as a pass-through for the electorate. Policy choices cannot
always (or even often) be reduced to the simplified model that allows a
median voter to emerge; there may be multiple dimensions along which
choices must be made such that no clear winner emerges from a majority
vote.194 The median-voter model (and majority choice in general) does not
account for the strength of preferences, and there are normative models of
190. The presidential representation hypothesis is functionally very similar to the “faction
reduction” argument for presidential power that is also commonly made. See Bressman, supra note 67,
at 491; Cooper & West, supra note 49, at 871; Christopher C. DeMuth, A Strategy for Regulatory
Reform, REG.: AEI J. ON GOV’T & SOC’Y, Mar.–Apr. 1984, at 25, 27 (the President’s “own
staff . . . evaluate[s] each regulation, relatively free of the interest-group pressures faced by the
regulatory agencies”); see also Livermore & Revesz, supra note 34, at 1344–50 (surveying capturebased justifications for presidential influence over agencies).
191. Anthony Downs, An Economic Theory of Political Action in a Democracy, 65 J. POL.
ECON. 135, 139 (1957). Alternatively, the policy goal might be to maximize a social welfare function,
based on the preferences of the citizenry, with political oversight by an electorally accountable
President as the best means to steer agencies in that direction. See, e.g., Bressman, supra note 67, at 490
n.145 (noting that cost-benefit analysis, a major feature of centralized regulatory oversight, “can be
understood to ask agencies directly to consider popular preferences in policymaking”); DeMuth &
Ginsburg, supra note 49, at 1081 (arguing that there is a connection between the President’s preferences
and efficiency criteria). See generally MATTHEW D. ADLER, WELL-BEING AND FAIR DISTRIBUTION:
BEYOND COST-BENEFIT ANALYSIS (2012) (defending a particular version of the social welfare function
that is tied to actual citizen preferences, but more loosely than traditional cost-benefit analysis).
192. This basic model is highly simplified, and the existence of multiple dimensions of
preferences for the slate of proposals can result in no clear winner in a set of pairwise votes. See
generally Peter J. Coughlin, Single-Peaked Preferences and Median Voter Theorems, in 2 THE
ENCYCLOPEDIA OF PUBLIC CHOICE 524 (Charles K. Rowley & Friedrich Schneider eds., 2004).
193. Adler refers to a policy bearing the “Plebiscitary Feature” if it “would be chosen by a
majority of the citizens in some kind of hypothetical plebiscite.” Matthew D. Adler, Judicial Restraint
in the Administrative State: Beyond the Countermajoritarian Difficulty, 145 U. PA. L. REV. 759, 789
(1997). Adler argues that this concept undergirds claims of a countermajoritarian difficulty of both
statutes and regulation. Id. at 788–96.
194. See generally Kenneth J. Arrow, A Difficulty in the Concept of Social Welfare, 58 J. POL.
ECON. 328 (1950).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
83
policy-making that require departure from median preferences when they
conflict with individual rights or undermine aggregate well-being. 195
Nevertheless, a rough concept of a median voter provides a useful
simplification of the presidential representation hypothesis. Certainly, some
connection between the President and the preference of the electorate is
required to support the presidential representation hypothesis.196 The
presidential representation hypothesis is meant to legitimate presidential
power, and some relationship to democratic values is necessary. The
median-voter model provides a first approximation to evaluate presidential
representativeness.197 To the extent that presidents are expected to
systematically diverge from the preferences of the median voter,
presidential-control skeptics would be warranted in demanding an
explanation that accords with some normative vision of democratic
responsiveness.198
The median-voter model also helpfully provides a mechanism to
connect the normative goal—policies that approximate those that would be
chosen through majority vote—and the means for achieving it—agency
responsiveness to a President who is selected through majority vote. If the
hypothesis is correct, then the President’s policy choices will generally sit
toward the middle of electoral preferences, with roughly equal percentages
of voters on either side, because that is how presidents win elections.
Enhancing the influence of the President would thus bring agency policies
toward the center.
There are many potential ways for this mechanism to break down,
driving presidential preferences away from the median voter in the national
electorate. As an initial matter, presidents are not selected through a general
195. A standard social welfare function can generate different results than a simple majority vote
on policy choices. See generally ADLER, supra note 191. For a less preference-based account, see
AMARTYA SEN, DEVELOPMENT AS FREEDOM (1999).
196. This is especially the case if the majoritarian character of the President is meant to
counteract the countermajoritarian difficulty faced by courts and agencies. See Bressman, supra note
67, at 464–65. An alternative “transformative” conception of the presidency is not quite as tied to
existing preferences as a median-voter model, but nonetheless remains linked to desires within the
polity. See Stephen Skowronek, Twentieth-Century Remedies, 94 B.U. L. REV. 795, 801 (2014).
197. Nicholas O. Stephanopoulos, Elections and Alignment, 114 COLUM. L. REV. 283, 287
(2014) (“[A]n impressive range of thinkers have made the normative argument that the preferences of
voters ought to be aligned with those of their representatives—and that elections are the key instrument
for producing this alignment.”).
198. ROBERT A. DAHL, POLYARCHY: PARTICIPATION AND OPPOSITION 1 (1971) (“[A] key
characteristic of a democracy is the continuing responsiveness of the government to the preferences of
its citizens, considered as political equals.”). Presidents and other legal actors are also, either
intentionally or incidentally, involved in preference formation. For discussion of resulting complexities
for democratic theory, compare John Ferejohn, Must Preferences Be Respected in a Democracy?, in
THE IDEA OF DEMOCRACY 231 (David Copp et al. eds., 1993), with Cass R. Sunstein, Democracy and
Shifting Preferences, in THE IDEA OF DEMOCRACY, supra, at 196.
2 LIVERMORE 45-134 (DO NOT DELETE)
84
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
nationwide popular vote, and the Electoral College could create incentives
for presidents to favor certain “swing” regions of the country. In their
forthcoming book, The Particularistic President, Douglas Kriner and
Andrew Reeves examine how presidents focus their attention on electorally
important sections of the country, using data on “federal grant spending,
natural disaster declarations, and military base closures to show the
powerful influence of particularistic impulses on presidential
policymaking.”199 What they find is that “the policy consequences of this
particularism are stark, in many cases producing inequalities in the
allocation of federal resources that exceed those produced by Congress.”200
Regional electoral incentives, then, can drive presidential allocation and
policy choices away from the preferences of the median voter.
The system of private campaign finance may drive presidents to pursue
the median donor rather than the median voter. Presidential campaigns are
enormously expensive; spending on both the 2008 and 2012 elections
topped $2.5 billion.201 A tiny fraction of Americans provide the lion’s share
of donations; 0.53% of the U.S. voting-age population made a political
contribution of more than $200 in 2012, but donors in that category
accounted for more than 60% of total campaign contributions.202
Unsurprisingly, research has found that donors to political campaigns are
not drawn randomly from the population; they are wealthier and tend to
have distinct policy views.203
Concerns about the intersection of income inequality and political
participation have led to a sustained research effort to examine the
relationship between the two. In 2004, a Task Force on Inequality and
American Democracy was convened by the American Political Science
Association (APSA) to examine the question. The APSA ultimately
released a report and edited volume that raise concerns that “[t]he political
playing field [is] highly unequal.”204 Using data on congressional voting
199. Douglas Kriner & Andrew Reeves, The Electoral College and Presidential Particularism,
94 B.U. L. REV. 741, 752 (2014) (describing the book).
200. Id.
201.
The Money Behind the Elections, OPENSECRETS.ORG, http://www.opensecrets.org/
bigpicture/ (last visited Sept. 25, 2015).
202.
Donor Demographics, OPENSECRETS.ORG, http://www.opensecrets.org/bigpicture/
donordemographics.php?cycle=2012 (last visited Sept. 25, 2015).
203. See Stephanopoulos, supra note 197, at 339 (citing studies showing differences in donor
preferences compared to the population as a whole).
204. Lawrence R. Jacobs & Theda Skocpol, American Democracy in an Era of Rising Inequality,
in INEQUALITY AND AMERICAN DEMOCRACY: WHAT WE KNOW AND WHAT WE NEED TO LEARN 1, 12
(Lawrence R. Jacobs & Theda Skocpol eds., 2005); see TASK FORCE ON INEQUALITY AND AM.
DEMOCRACY, AM. POLITICAL SCIENCE ASS’N, AMERICAN DEMOCRACY IN AN AGE OF RISING
INEQUALITY (2004), http://www.apsanet.org/portals/54/Files/Task%20Force%20Reports/taskforce
report.pdf.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
85
records and other public policy outputs as well as survey responses within
the general population, researchers have found that politicians are relatively
more responsive to wealthy constituents.205 Certain research has found
relatively little divergence on policy matters by income,206 but studies that
have specifically targeted the wealthiest Americans have found stark
differences on policy positions as well as substantially greater levels of
political participation.207 The most recent research, which uses more
extensive data on policy attitudes within the population, finds substantially
greater policy responsiveness to the preferences of the wealthy, especially
within the Republican Party.208
Most relevant for the following discussion is the role of two-stage
elections, in which candidates first vie for the nomination of one of the two
major political parties, and then proceed to a general election. Under some
models of political parties in which electoral success is the ultimate end,
parties gravitate toward candidates who reflect median preferences.209 But
even when electoral success is paramount, the need to raise funds may give
outsized influence to donors,210 and politicians may additionally benefit
from distinctive policy “brand,” which requires differentiation and some
205. See LARRY M. BARTELS, UNEQUAL DEMOCRACY: THE POLITICAL ECONOMY OF THE NEW
GILDED AGE (2010); MARTIN GILENS, AFFLUENCE AND INFLUENCE: ECONOMIC INEQUALITY AND
POLITICAL POWER IN AMERICA (2012); Lawrence R. Jacobs & Benjamin I. Page, Who Influences U.S.
Foreign Policy?, 99 AM. POL. SCI. REV. 107 (2005); see also SIDNEY VERBA, KAY LEHMAN
SCHLOZMAN & HENRY E. BRADY, VOICE AND EQUALITY: CIVIC VOLUNTARISM IN AMERICAN POLITICS
(1995) (discussing income-, wealth-, and education-related disparities in participation in “civic
voluntarism”).
206. See Yosef Bhatti & Robert S. Erikson, How Poorly Are the Poor Represented in the U.S.
Senate?, in WHO GETS REPRESENTED? 223 (Peter K. Enns & Christopher Wlezien eds., 2011); Stuart
N. Soroka & Christopher Wlezien, On the Limits to Inequality in Representation, 41 PS: POL. SCI. &
POL. 319 (2008); Joseph Daniel Ura & Christopher R. Ellis, Income, Preferences, and the Dynamics of
Policy Responsiveness, 41 PS: POL. SCI. & POL. 785 (2008).
207. See Benjamin I. Page, Larry M. Bartels & Jason Seawright, Democracy and the Policy
Preferences of Wealthy Americans, 11 PERSP. ON POL. 51 (2013).
208. See Jesse H. Rhodes & Brian F. Schaffner, Economic Inequality and Representation in the
U.S. House: A New Approach Using Population-Level Data (Apr. 7, 2013) (unpublished manuscript)
(on file with the Alabama Law Review).
209. Downs, supra note 191.
210. See David P. Baron, Electoral Competition with Informed and Uniformed Voters, 88 AM.
POL. SCI. REV. 33 (1997) (describing conditions that will lead to divergence from median preferences
and toward donor preference). See also Timothy Feddersen & Faruk Gul, Polarization and Income
Inequality: A Dynamic Model of Unequal Democracy (2014) (unpublished manuscript) (showing
polarization resulting from a need for policy-oriented politicians to compete for both donors and votes,
in conditions of economic inequality and income-based policy divergence). See generally KAY LEHMAN
SCHLOZMAN, SIDNEY VERBA & HENRY E. BRADY, THE UNHEAVENLY CHORUS: UNEQUAL POLITICAL
VOICE AND THE BROKEN PROMISE OF AMERICAN DEMOCRACY (2012) (discussing interaction of
campaign finance, interest groups, and parties on political representation); THOMAS FERGUSON,
GOLDEN RULE: THE INVESTMENT THEORY OF PARTY COMPETITION AND THE LOGIC OF MONEYDRIVEN POLITICAL SYSTEMS (1995) (arguing that political parties represent left and right branches of
coalition of moneyed elites that vie for political office).
2 LIVERMORE 45-134 (DO NOT DELETE)
86
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
distance from median preferences.211 Still other models view electoral
success as merely the means to the end of achieving benefits for members
of the party coalition. Given this different set of goals, parties will use the
nomination process to screen for candidates who will deliver those benefits,
subject to the constraint of gaining a bare majority on election night.
These competing models of the presidential nomination process have
obvious implications for the presidential representation hypothesis. If the
former, electoral-success model is correct, then presidents can be expected
to occupy the center on the political spectrum. If the latter, party-coalitionbenefits model is more accurate, then presidents will instead reflect the
narrow interests of the party coalition.
These competing models of the presidential nomination process have
obvious implications for the presidential representation hypothesis. Under
some, presidents can be expected to occupy the center on the political
spectrum, and the presidential representation hypothesis has some validity.
If other alternatives are more accurate, then presidential preferences will
not occupy the political center, and the standard justification for
presidential oversight breaks down. With the decline of traditional party
organizations and the rise of contemporary parties discussed in Part I.C,
political scientists began to ask which model better characterizes
contemporary presidential nomination politics. It is to their competing
answers to this question that this Article now turns.
B. Politicians and Groups in Political Parties
This section evaluates the presidential representation hypothesis in
light of current thinking within the political science literature on the role of
parties in the nomination process. The conclusion is that the presidential
representation hypothesis finds little support, even under the most
favorable interpretation of contemporary parties. If a strong role for the
President in agency oversight is justified, it must find alternative normative
foundations.
For Aldrich, the 1960 presidential election marked the dawn of
candidate-centered elections, in which the resources of traditional party
organizations were no longer necessary to mount a successful presidential
bid. But if John F. Kennedy proved that candidates could create the
campaign operations needed to win a general election, it was Jimmy
Carter’s successful contest for the 1974 Democratic nomination that raised
the question of whether party insiders were able to exercise any substantial
influence over even the nomination process for the single most important
211. See generally Aldrich, supra note 6.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
87
elected office in the country. The seeming irrelevance of party insiders in
the primary-driven presidential nomination process contributed to the view
that the era of party politics had been supplanted by one in which elections
would center on the candidates themselves.212
An era of candidate-centered elections is consistent with a model of
politician-centered parties, in which electoral success is the ultimate goal of
the party. In his highly influential rational-choice model of politics,
Anthony Downs defines a party as a “team . . . who seek[s] office solely in
order to enjoy the income, prestige, and power that go with running the
governing apparatus.”213 Such parties “do not seek to gain office in order to
carry out certain preconceived policies[,] . . . rather they formulate
policies . . . to gain office.”214 This understanding of parties led Downs to
the median-voter conclusion—parties that are exclusively concerned with
holding office will seek to maximize their share of total votes, and will do
so by locating themselves at the preferred preferences of the median
voter.215
Joseph A. Schlesinger provided an important extension on Downs’s
work by elaborating on how parties coordinate the members of the “team,”
recognizing the importance of “benefits seekers” that provide inputs into
electoral success. These benefits seekers have different electoral goals, and
they attempt to pursue them by pushing the party toward a minimum
winning coalition, rather than by maximizing vote shares.216 The tension
between the benefits seekers and politicians arises when “the two goals []
impose conflicting views of how to win elections and, ultimately,
conflicting views of how parties should be organized.”217 Although
Schlesinger recognizes the importance of benefits seekers, he ultimately
“assert[s] that no matter how intense the struggle, the goal of office and the
office seeker will ultimately prevail.”218 As a consequence, parties will
generally attempt to maximize vote shares through appeals to the political
center. Demsetz describes a similar structure of balancing the preferences
of “internal constituencies” (i.e., party leaders, members, and supporters)
212. MARTIN P. WATTENBERG, THE DECLINE OF AMERICAN POLITICAL PARTIES, 1952–1996
(5th ed. 1998).
213. Downs, supra note 191, at 137.
214. Id.
215. Id.
216. JOSEPH A. SCHLESINGER, POLITICAL PARTIES AND THE WINNING OF OFFICE 157 (1991).
217. Id. at 148.
218. Id. Schlesinger also provides an account of “centrifugal” and “centripetal” pressures on
political parties, essentially arguing that party coordination will decline (centrifugal behavior) when the
benefits of incumbency dominate and electoral competition declines, while party coordination
(centripetal behavior) will increase as electoral competition increases. Schlesinger argues that changes
in the electorate mid-century switched incentives in ways that led to increased centripetal pressure,
contributing to more coordinated parties after the 1960s. See id. at 177–99.
2 LIVERMORE 45-134 (DO NOT DELETE)
88
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
and external constituencies (i.e., voters).219 But Demsetz concludes that
parties will not seek to maximize vote share and drift away from the
political center.220
Group-centered theories emphasize the benefits seekers and view
parties as primarily collections of interest groups that form coalitions to use
influence over the electoral process to effectuate policy change. Influenced
by pluralistic political theory, an earlier version of the group-centered
approach was prominent during the New Deal period but fell out of favor
for several decades.221 The transition of parties in the past twenty-five years
has spurred a recent renaissance.222 In one account, contemporary parties
are composed of “intense policy demanders” made up of “interest groups
and activists” who work in coalition to
develop[] an agenda of mutually acceptable policies, insist[] on the
nomination of candidates with a demonstrated commitment to [the
party] program, and work[] to elect these candidates to office. In
this group-centric view of parties, candidates will, if the coalition
has selected them well, have as their paramount goal the
advancement of the party program.223
The group-centered approach paints a less “rosy” picture of the role of
parties in a democratic system.224 Rather than forward politicians’ goals of
maximizing electoral success—which ultimately translates into
responsiveness to voter demands—parties are meant to help interest groups
“capture and use government for their particular goals” through any (legal)
means necessary, including by taking advantage of voter inattention.225
Under group-centered accounts, intense policy demanders impose
discipline on politicians through the primary process and by cultivating
candidates who have strong personal commitments to the party program.
219. See id. at 149–51.
220. Harold Demsetz, Amenity Potential, Indivisibilities, and Political Competition, in
PERSPECTIVES ON POSITIVE POLITICAL ECONOMY 144, 149 (James E. Alt & Kenneth A. Shepsle eds.,
1990).
221. See, e.g., KEY, supra note 124.
222. See COHEN ET AL., supra note 10.
223. Bawn et al., supra note 18, at 571, 591.
224. Id. at 571. Note that the group-centered descriptive account is different from normative
models from a different time, which accentuated the role of parties in facilitating pluralistic
representation through interest group bargaining. Cf. KEY, supra note 124. Parties dominated by intense
policy demanders are unlikely, as an empirical matter, to act as conduits for genuine pluralistic
representation because only some factions will be strongly represented. Cf. COHEN ET AL., supra note
10; Bawn et al., supra note 18.
225. Id.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
89
This second mechanism would be effective even for presidents, who are
unlikely to face primary challenges once elected.
Aldrich’s account of parties attempts to harmonize a rational-choice,
politician-centered understanding of political parties with the role of
benefits-seeking activists in contemporary parties and a more realistic
behavioral account of the motivations for office-seeking.226 Part of what
drives party maintenance in an era of candidate-centered elections is the
value of a party brand that signals a particular set of programmatic
positions. This signal is only valuable to the extent that it is coherent,
understandable by the public, and distinct from the other party. When this
signaling function is important, parties tend to “polarize” by adopting
distinct programs that are used to recruit politicians and engage the activist
base of volunteers and donors. The party signal is a public good that is
shared by all affiliated politicians, and the contemporary party is a
mechanism to organize those politicians to produce and amplify that signal,
in light of collective-action challenges.227
An additional complication for the presidential representation
hypothesis is growing partisanship within the electorate. The Downsian
median-voter model anticipates a relatively normal distribution of political
preferences along a common dimension, in which there is a relatively large
group of people in the middle with moderate preferences, and diminishing
populations toward the extremes. Some recent research suggests that
partisanship within the electorate has resulted in a distribution of
preferences with a Republican distribution on the right, a Democratic
distribution on the left, and a relatively small population in the center.228 If
the electorate is divided in this way, and has some degree of partisan
loyalty, it is not clear that reaching toward the center will be the most
successful electoral strategy. Others argue that polarization is largely an
elite phenomenon, and that there remains a large center within the
American electorate.229
It is unlikely that contemporary parties are either purely group-centered
or politician-centered. Rather, a party at a given time might be
characterized as relatively more group- or politician-centered, depending
on the temporary allocation of power, even as neither politicians nor groups
are entirely shut out. Allocation of power within a party between politicians
and groups will be affected by institutional arrangements (such as the
226. See generally ALDRICH, supra note 6.
227. Id.
228. See, e.g., ALAN I. ABRAMOWITZ, THE POLARIZED PUBLIC: WHY AMERICAN GOVERNMENT
IS SO DYSFUNCTIONAL (2012).
229. See, e.g., MORRIS P. FIORINA & SAMUEL J. ABRAMS, DISCONNECT: THE BREAKDOWN OF
REPRESENTATION IN AMERICAN POLITICS (2009).
2 LIVERMORE 45-134 (DO NOT DELETE)
90
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
structure of primaries), communications technologies, fundraising rules,
and the charismatic qualities of individual politicians.230
One feature of contemporary parties that are not well captured by the
two dominant accounts is the outsized role of one particular politician—the
President. In addition to sitting at the head of the Executive Branch,
presidents are the functional leaders of their parties while in office, and
typically continue to engage in substantial party-building activities outside
of office. Presidents bring different styles of partisanship to their jobs, and
tend to leave lasting imprints on their parties.231 Realistically, there might
be thought to be three distinct centers of gravity within parties: benefitseeking interest; office-seeking politicians; and presidents. The latter center
of gravity is, perhaps, the most difficult to understand because presidents
are individuals rather than collectives—idiosyncratic motivations and
personal style are less likely to be swamped by systematic pressures.
From the perspective of political control over agencies, when parties
are relatively more group-centered, presidential oversight powers will be
used to favor agency decisions that depart from those of the median
voter.232 Even more politician-centered parties, which may select
candidates who are relatively more responsive to voters and who will be
relatively more likely to use presidential oversight to push agencies toward
the center, are unlikely to truly track median voter preferences.233
Presidentially centered parties will respond to the personal preferences of
the President, which may be oriented to voters, partisans, or historical
legacy, depending on the temperament of the person in the White House.
None of these theories of parties, then, well-accords with the presidential
representation hypothesis.
230. Andrea Louise Campbell, Parties, Electoral Participation, and Shifting Voting Blocs, in
THE TRANSFORMATION OF AMERICAN POLITICS: ACTIVIST GOVERNMENT AND THE RISE OF
CONSERVATISM 68, 68 (Paul Pierson & Theda Skocpol eds., 2007).
231. STEPHEN SKOWRONEK, THE POLITICS PRESIDENTS MAKE: LEADERSHIP FROM JOHN ADAMS
TO BILL CLINTON (2d ed., 1997); Sidney M. Milkis, Jesse H. Rhodes & Emily J. Charnock, What
Happened to Post-Partisanship? Barack Obama and the New American Party System, 10
PERSPECTIVES ON POLITICS 57 (2012).
232. Of course, if administrative policy outputs are too extreme, then the party may suffer
electoral consequences, so voters act as a check on even group-centered parties.
233. Kagan acknowledged that the normative case for presidential control is weak if presidents
are pulled away from majority preferences by party demands. See Kagan, supra note 3. To counter that
argument, she includes a citation to “scholarly commentary on past Presidents’ relative lack of
partisanship.” See id. at 2311 n.261 (citing Milkis, supra note 115); Austin Ranney, The President and
His Party, in BOTH ENDS OF THE AVENUE: THE PRESIDENCY, THE EXECUTIVE BRANCH AND CONGRESS
IN THE 1980S, at 131 (Anthony King ed., 1983).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
91
One strategy to test how theory matches reality is to examine the
nomination process.234 Truly politician-centered parties would acquiesce to
the strongest candidate (i.e., the candidate who best tracks median
preferences), while group-centered parties screen candidates on the basis of
loyalty to groups in the coalition, even at the risk of electoral defeat.235 In a
significant study of presidential primaries in the contemporary period,
Cohen, Karol, Noel, and Zallerfind find that a group-centered theory of
presidential nominations is better supported.236
The Cohen et al. study takes as its starting place what the authors refer
to as the “invisible primary,” which occurs prior to any public vote.237 The
function of this invisible primary is to simulate the insider nomination
process that existed prior to the reforms in the nomination process in the
wake of the 1968 Democratic convention. In the invisible primary, intense
policy demanders operate through informal party networks to screen
potential presidential candidates for acceptability. Although electability is
one of the criteria applied to candidates, conformity with the policy goals
of groups within the party coalition is at least as important, and groups
trade electoral risk against candidate fidelity to their interests.
Because the invisible primary is difficult to observe directly, Cohen et
al. use endorsements by high-profile party figures—namely prominent
elected officials—as a proxy for this process. The authors do not assume
that endorsements themselves operate to effectively screen candidates, but
instead that prominent endorsements are responsive to the same general
forces that operate in the invisible primary. Gathering data on pre-primary
endorsements in the period from 1980 to 2008, they find that in the vast
majority of cases, the candidates with the largest number of endorsements
going into the primary season were the party’s ultimate nominee. Based on
this data, they conclude that “[r]ank-and-file voters possess the formal
power to nominate, but they normally follow the insider consensus.”238
One potentially interesting wrinkle in this data is that more recent
presidential elections, especially on the Democratic side, break this trend.
Both John Kerry and Barack Obama enjoyed substantially less support
from Democratic Party insiders in the lead-up to the primary season, but
234. See SETH E. MASKET, NO MIDDLE GROUND: HOW INFORMAL PARTY ORGANIZATIONS
CONTROL NOMINATIONS AND POLARIZE LEGISLATURES (2009) (arguing that informal organizations
have an extremely important role in nominations).
235. For contemporary examples of willingness to risk losses in general elections to promote
adherence to policy programs, see THEDA SKOCPOL & VANESSA WILLIAMSON, THE TEA PARTY AND
THE REMAKING OF REPUBLICAN CONSERVATISM (2012).
236. COHEN ET AL., supra note 10.
237. Much of their analysis is based on data concerning public endorsements of primary
candidates by elected officials.
238. Bawn et al., supra note 18, at 586 (summarizing COHEN ET AL., supra note 10).
2 LIVERMORE 45-134 (DO NOT DELETE)
92
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
both emerged as the party’s eventual candidate. Indeed, both lagged behind
two other candidates in early endorsements (Kerry behind Gephardt and
Dean; Obama behind Clinton and Edwards). John McCain was also
deemed unacceptable to a substantial percentage of Republican Party
insiders but gained the nomination in 2008.239 Distrust of Mitt Romney was
also widespread within the Republican Party, and although he eventually
captured the nomination, it was only after a costly battle. These recent
examples in both parties indicate that the trend identified by Cohen et al. of
strong insider influence over the nomination process may be fading.
Other lines of research also tend to indicate that presidents are not well
described as proxies for the median voter. In an extensive study of
presidential behavior, B. Dan Wood codes “every sentence spoken publicly
by the president” from President Truman through the end of the first term
of President George W. Bush along nine issue domains according to a
conservative and liberal axis.240 The relative liberalness of presidential
statements are then compared to polling data on public sentiment to test
how well the centrist model explains the public positions of sitting
presidents. Wood finds that “presidents do not cater systematically to mass
preferences” but instead “consistently express partisan issue stances.”241 He
further finds that presidents are “unresponsive to incentives from changing
public approval and elections” and that the main determinants of the
liberalisms of presidential statements are “partisanship . . . tempered by
pragmatic concerns,” including the economic condition of the country and
the partisan composition of Congress.242 Wood’s finding provides support
for a group-centered model in which presidents cater to their base, rather
than reach for the center. The expectation of partisan behavior by
presidents may help explain why “off-side plays”—where presidents adopt
policies at odds with their party program—are such “effective maneuvers
when deployed.”243 The relative rarity of such moves attests to the general
regularity of presidential conformity to their partisan commitments.
239. McCain’s nomination was made possible by particularly strong showings in New
Hampshire and South Carolina, which have open primaries. Open primaries have some constitutional
limits. See Wash. State Grange v. Wash. State Republican Party, 552 U.S. 442 (2008) (upholding a
“modified blanket primary” that allowed a candidate to affiliate with the party of his choosing on a
general ballot); Cal. Democratic Party v. Jones, 530 U.S. 567 (2000) (invalidating California’s blanket
primary system as violation of freedom of association).
240. B. DAN WOOD, THE MYTH OF PRESIDENTIAL REPRESENTATION 67 (2009).
241. Id. at 118.
242. Id. at 118–19. See also Kevin Coe, The Language of Freedom in the American Presidency,
1933-2006, 37 PRESIDENTIAL STUD. Q. 375 (2007) (noting partisan difference in how presidents discuss
the concept of “freedom”).
243. Robert E. Goodin, Voting Through the Looking Glass, 77 AM. POL. SCI. REV. 420 (1983).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
93
Overall, there is very little theoretical and empirical support in the
literature on contemporary parties for the presidential representation
hypothesis. Under group-centered accounts of parties, presidents will use
their oversight powers to promote the interests of activist party
constituencies with extreme policy preferences. Even under politiciancentered accounts, the need to build a party brand and respond to
partisanship within the electorate will drive presidents away from the
preferences of median voters. Available evidence is consistent with these
theoretical predictions. In the main, party insiders play an important role in
selected presidential nominees, and once in office, presidents appear to take
positions that are consistent with their party programs, rather than
occupying the political center. Given the weakness of the presidential
representation hypothesis, it provides an inadequate justification for the
strong role currently given to presidents to oversee the administrative state.
C. Responsible Party Government
A simple median-voter proxy model of presidential representation as a
defense for presidential administration is unsatisfying. At the very least, the
Electoral College and the system of private campaign finance potentially
distort presidential priorities toward particular regions or socio-economic
groups. More centrally for the analysis in this Article, the two-party
system—coupled with parties that are at least somewhat responsive to
internal constituencies rather than voters—implies that presidents will
adopt positions at some distance from the median voter.244 Perhaps on
average the center is represented, but one is reminded of the joke about the
statistician with his head in an oven and his feet in a bucket of ice; when
asked how he felt, he said: “On average, I feel just fine.”245
The era of candidate-centered elections has not solved this problem.
The legal, technological, and cultural changes that led to the breakdown of
traditional party organizations has not resulted in purely politiciandominated parties that seek to maximize their vote share by tracking
median preferences. Party insiders appear to continue to exercise at least
some degree of control over presidential nominations, although the extent
of their influence is subject to debate. Presidents, at least in their public
statements, do not appear simply to track public preferences, but instead
244. Matthew C. Stephenson, Optimal Political Control of the Bureaucracy, 107 MICH. L. REV.
53, 55 (2008) (arguing that some degree of insulation may be justified).
245. Dr. Gary C. Ramseyer, Sir Ronald A. Fisher Hall, ILL. ST. U., http://my.ilstu.edu/
~gcramsey/FisherHall.html (last visited Sept. 25, 2015) (describing the joke as the “granddaddy” of
statistics humor).
2 LIVERMORE 45-134 (DO NOT DELETE)
94
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
represent the mainstream view of their parties. Any satisfying justification
for presidential administration must fit with this reality.
A useful starting place for squaring presidential administration with the
realities of party politics is a normative understanding of political parties
and their role in democratic societies. Political scientists have long noted
the benefits of political parties: they facilitate the electoral process by
reducing the costs of voting, recruiting candidates, and encouraging voters;
they professionalize the political class by providing job security; they
increase capacity for governance, develop policy proposals, and coordinate
constitutionally fractured government authority; and they mediate conflict
between existing social groups and create opportunities for outsiders to
gain influence.246 Without political parties, it is unclear how the basic
machinery of elections and governing would be accomplished. From a
functional standpoint, political parties are basic components of a workable
democratic society.
One articulation of the normative role for parties in democratic
societies is the “responsible party government” account, described midtwentieth century by E.E. Schattschneider and an American Political
Science Association committee that he chaired.247 The basic components of
responsible party government are two parties that are genuinely
competitive, are sufficiently different to provide voters with a choice, and
that announce policy programs and carry them out while in office or
criticize the party in power when out of office.248 Under this model, parties
are able to overcome some of the stagnation and gridlock inherent in the
American separation-of-powers system and play an essential role in
facilitating democratic accountability by developing slates of policy
proposals, putting them up for a vote, and standing on their records of
accomplishment.249 For Aldrich, contemporary political parties are well
suited to carrying out these tasks, and he has argued that in their current
form, American parties have come close to “a nearly mythical version of
Westminster-style democracy akin to many of the aspects championed by
those in favor of the responsible party thesis.”250
246. See SCHUMPETER, supra note 93, at 250–68.
247. See generally E.E. SCHATTSCHNEIDER, PARTY GOVERNMENT (1942); COMM. ON POLITICAL
PARTIES OF THE AM. POLITICAL SCI. ASS’N, TOWARD A MORE RESPONSIBLE TWO-PARTY SYSTEM
(1950).
248. COMM. ON POLITICAL PARTIES OF THE AM. POLITICAL SCI. ASS’N, supra note 247, at 1–2.
249. See generally AUSTIN RANNEY, CURING THE MISCHIEFS OF FACTION: PARTY REFORM IN
AMERICA (1975). For an assessment fifty years later of the work of the Committee for a More
Responsible Two-Party system, see CONFERENCE PROCEEDINGS — APSA RESPONSIBLE PARTIES
PROJECT, 1950–2000, http://www.apsanet.org/~pop/proceedings.htm.
250. See ALDRICH, supra note 6, at 296.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
95
Other normative accounts of parties have emphasized their role in
facilitating democratic action in a pluralistic society. By competing for
votes from different interest groups, aggregating interests within the party,
and facilitating negotiation and compromise between those interests, parties
serve as a pluralistic forum and mechanism for the articulation and
vindication of collective goals. Along these lines, Professor Rosenblum
defends party politics as encouraging the moral values of “inclusiveness,
comprehensiveness, and disposition to compromise” that are fundamental
for successful pluralistic democracies.251 These values “are congruent with
standard democratic virtues” such as “mutual respect, minimal concern for
the interests and opinions of others, provisionality, and resolving disputes
through argument.”252 But party rivalry specifically requires partisans to
“identif[y] with others in [a] system of regulated rivalry,”253 whereby no
party “speak[s] for the whole” but partisans “think[] they should speak to
everyone.”254 This system of party rivalry requires “commitment to the
provisional nature of political authority, its periodic recreation . . . [that] is
the distinguishing feature of representative democracy, and the moral
distinctiveness of partisanship.”255
Extending these normative visions of contemporary parties (which are
at least potentially consistent with empirical reality) to presidential
administration, a different, less majoritarian justification emerges.
Presidential power is not justified for its potential to bring agencies in line
with median preferences. This justification founders on the shoals of
partisan reality. Instead, presidential oversight facilitates responsible party
governance by creating a means for the policy programs that were
developed by the winning party to be implemented in practice. This fulfills
one of the basic requirements of responsible party government—that
parties be “willing and able to carry . . . out [policy commitments] when in
office.”256 Presidential control also facilitates attribution of the choices
made by agencies to an elected official. Cutting agencies loose from
presidential influence undermines the ability of parties to affect policy by
winning elections and the ability of voters to hold an elected official
251. ROSENBLUM, supra note 91, at 362–64. “Among political identities, only partisanship has
this potential for inclusiveness, comprehensiveness, and disposition to compromise.” Id. at 362
(emphasis omitted). It is worth noting that Rosenblum is very interested in the felt identification with
party, which is emphasized less in Cohen et al. and Aldrich. Hers is a “party-in-the-electorate” oriented
theory, while Cohen et al. and Aldrich focus on elites.
252. Id. at 362.
253. Id. at 362.
254. Id. at 365.
255. Id. at 363.
256. NELSON W. POLSBY & AARON B. WILDAVSKY, PRESIDENTIAL ELECTIONS: STRATEGIES OF
AMERICAN ELECTORAL POLITICS 208 (4th ed. 1976) (emphasis added).
2 LIVERMORE 45-134 (DO NOT DELETE)
96
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
accountable for agency actions. Only if “elections have consequences”257
can responsible parties emerge and partisans be given an incentive to
cultivate Rosenblum’s pluralistic qualities of “inclusiveness,
comprehensiveness, and disposition to compromise.”258
Understanding the virtue of the presidency in programmatic rather than
representational terms may be the beginning of a new normative account of
presidential oversight, but it is not the end. Presidential influence can
facilitate a dynamic interplay between parties that have incentives to
develop distinct programs, test them for voter appeal during elections, and
implement them while in power. But Congress also has a stake in agency
oversight, and the collision of contemporary political parties and the
constitutional structure of separation of powers creates complications for
responsible party government. The promise and pitfalls of the legislativeexecutive split are discussed in the next Part. In addition, administrative
law embeds other social values, including neutral expertise, vigor,
coherence, and legality that are no easy fit with party rivalry. In Part V, this
Article will turn to the question of how to reconcile these values with the
benefits brought about by responsible party government.
IV. RELATIONSHIPS WITH CONGRESS
Congress plays a substantial role in overseeing the Executive Branch of
government in the United States.259 This power stems both from Congress’s
position in the constitutional structure and from steps that Congress itself
has taken to increase its internal capacity for executive oversight. The
shared oversight dynamic between Congress and the President raises a host
of questions that have occupied political scientists since the dawn of the
field,260 including the comparative institutional incentives for the President
257. This phrase is frequently attributed to Dick Gephardt. See Leo Hindery, Jr., Elections Have
Consequences—And Carry Messages, HUFFINGTON POST (Nov. 12, 2012, 9:30 AM),
http://www.huffingtonpost.com/leo-hindery-jr/elections-have-consequenc_b_2116585.html.
258. ROSENBLUM, supra note 91, at 362. Jerry Mashaw makes a similar point by raising
concerns that in a world where agency decisions are divorced from presidential preferences, elections
would be reduced to “beauty contests.” See Mashaw, supra note 188, at 96.
259. See JOEL D. ABERBACH, KEEPING A WATCHFUL EYE: THE POLITICS OF CONGRESSIONAL
OVERSIGHT 20 (1990) (noting congressional oversight during early years of the Republic). See
generally MASHAW, supra note 35; CONGRESS INVESTIGATES: A DOCUMENTED HISTORY, 1792–1974
(Arthur M. Schlesinger, Jr. & Roger Bruns eds. 1975). Aberbach goes so far as to characterize the U.S.
Congress as “the world’s most powerful legislative body[,] with great influence over agency behavior.”
ABERBACH, supra, at 22.
260. See, e.g., WOODROW WILSON, CONGRESSIONAL GOVERNMENT (1885). For a more recent
entry, see Alan E. Wiseman, Delegation and Positive-Sum Bureaucracies, 71 J. POLITICS 998 (2009).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
11/5/2015 9:20 AM
Political Parties and Presidential Oversight
97
and Congress to carry out oversight activities261 and each institution’s
likely success in bringing agencies to heel.262
The rise of contemporary parties may influence Congress and the
relationship between the President and Congress in many ways. The
following discussion focuses on how contemporary parties affect the
dynamic of shared oversight and will focus on the potential for conflict
between the branches that arises from agency oversight. The programmatic,
national, and professionalized nature of contemporary parties affects the
ability of individual members of Congress to coordinate over shared goals
and the distribution of power in the body. The growth of contemporary
parties has likely helped shift the emphasis of oversight from memberconstituent-driven interests toward promotion of party programmatic goals.
Although some commentators have argued that this shift has exacerbated
tendencies toward gridlock that are built into the U.S. constitutional
system, there may be countervailing democratic benefits associated with
this form of public scrutiny of agency action that should be acknowledged
and considered.
A. Shared Oversight
Congress also has a variety of tools to influence administration. The
foundations of Congress’s influence over the administrative state are its
lawmaking power, its budgetary power, its role in appointment, and its
hearing and investigative power. Congress has built on this constitutional
foundation through internal organization to enhance its capacity to use
these powers effectively.
The legislative power is vested in the Congress, and as courts like to
say, administrative agencies are “creatures of statute, bound to the confines
of the statute that created them.”263 Although Congress sometimes provides
261. See Moe, supra note 12 (noting increasing expectations as a source of incentive for the
President).
262. See, e.g., John Ferejohn & Charles Shipan, Congressional Influence on Bureaucracy, 6 J.L.
ECON. & ORG. (SPECIAL ISSUE) 1 (1990) (investigating effect of Congress on bureaucracy); Thomas H.
Hammond & Jack H. Knott, Who Controls the Bureaucracy? Presidential Power, Congressional
Dominance, Legal Constraints, and Bureaucratic Autonomy in a Model of Multi-Institutional PolicyMaking, 12 J.L. ECON. & ORG. 119 (1996); Barry R. Weingast & Mark J. Moran, Bureaucratic
Discretion or Congressional Control? Regulatory Policymaking by the Federal Trade Commission, 91
J. POL. ECON. 765 (1983).
263. U.S. Fidelity and Guar. Co. v. Lee Invs. LLC, 641 F.3d 1126, 1135 (9th Cir. 2011) (citing
Int’l Union of Elec., Radio & Mach. Workers, AFL-CIO v. NLRB, 502 F.2d 349, 353 n.* (D.C. Cir.
1974) (“An administrative agency . . . is a creature of the statute that brought it into existence.”)).
2 LIVERMORE 45-134 (DO NOT DELETE)
98
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
wide grants of authority to agencies,264 the United States Code is full of
detailed statutory schemes that limit the field of action for agencies.265
Statutes govern not only regulatory ends and means, but also administrative
process and the structure of administrative agencies.266 Decisions
concerning the organization of the executive, including “what departments
to create, how to organize those departments into various authorities and
agencies[,] and whether to create agencies outside of any department,” rest
largely with Congress in the exercise of its lawmaking function.267
Congress has similarly expansive powers over the budget and frequently
uses appropriation riders to directly influence agency policy choices.268
264. For a critique of broad delegation to agencies, see SCHOENBROD, supra note 181; Theodore
J. Lowi, Two Roads to Serfdom: Liberalism, Conservatism and Administrative Power, 36 AM. U. L.
REV. 295 (1987).
265. See, e.g., Sierra Club v. EPA, 294 F.3d 155, 161 (D.C. Cir. 2002) (requiring the agency to
adhere to statutorily required deadlines for pollution reduction despite contrary policy considerations).
Of course, complexity can also breed discretion. See, e.g., Chevron, U.S.A., Inc. v. Natural Res. Def.
Council, 467 U.S. 837, 863–65 (1984) (justifying deference to the EPA Administrator in interpreting
Clean Air Act in part because the statute is “technical and complex”).
266. See STEVEN P. CROLEY, REGULATION AND PUBLIC INTERESTS: THE POSSIBILITY OF GOOD
REGULATORY GOVERNMENT (2008); Matthew D. McCubbins, Roger G. Noll & Barry R. Weingast,
Structure and Process, Politics and Policy: Administrative Arrangements and the Political Control of
Agencies, 75 VA. L. REV. 431 (1989). Statutory deadlines for rulemakings are another source of
congressional control. See William F. West & Connor Raso, Who Shapes the Rulemaking Agenda?
Implications for Bureaucratic Responsiveness and Bureaucratic Control, 23 J. PUB. ADMIN. RES. &
THEORY 495 (2013) (finding that nearly one-third of studied rules were specifically mandated by
Congress, with the remainder being discretionary exercise of statutorily created authority). Reporting
requirements are another congressional tool. See Jonathan G. Pray, Comment, Congressional Reporting
Requirements: Testing the Limits of the Oversight Power, 76 U. COLO. L. REV. 297 (2005). See
generally Lisa Schultz Bressman, Procedures as Politics in Administrative Law, 107 COLUM. L. REV.
1749 (2007) (arguing that the Supreme Court uses procedure to “negotiate” between congressional
control (procedure) and presidential control (deference)); Terry M. Moe & Scott A. Wilson, Presidents
and the Politics of Structure, LAW & CONTEMP. PROBS., Spring 1994, at 1, 39 (citing Robert V.
Percival, Checks Without Balance: Executive Office Oversight of the Environmental Protection Agency,
LAW & CONTEMP. PROBS., Autumn 1991, at 127, 175) (arguing that Congress throughout the 1980s
responded to Reagan’s imposition of presidential authority by “burying the EPA in more bureaucracy”).
267. Jack M. Beermann, Congressional Administration, 43 SAN DIEGO L. REV. 61, 107–08
(2006). Congress can use agency structure and decision-making processes to empower itself, entrench
favored interests, insulate agencies from presidential influence, or even set up an agency for failure. See
Moe & Wilson, supra note 266, at 6. See also CROLEY, supra note 266 (arguing that structure and
process can protect broad general interests); Barkow, supra note 64; Jonathan R. Macey,
Organizational Design and Political Control of Administrative Agencies, 8 J. L. ECON. & ORG. 93
(1992); McCubbins, Noll & Weingast, supra note 266.
268. See CURTIS W. COPELAND, CONG. RESEARCH SERV., RL 34354, CONGRESSIONAL
INFLUENCE ON RULEMAKING AND REGULATION THROUGH APPROPRIATIONS RESTRICTIONS (2008),
http://fas.org/sgp/crs/misc/RL34354.pdf; Jason A. MacDonald, Limitation Riders and Congressional
Influence over Bureaucratic Policy Decisions, 104 AM. POL. SCI. REV. 766, 767 (2010); Kate Stith,
Congress’ Power of the Purse, 97 YALE L.J. 1343, 1352–56 (1988); Hans J.G. Hassell & Samuel
Kernell, Veto Rhetoric and Legislative Riders (2013) (unpublished manuscript),
http://pages.ucsd.edu/~skernell/resources/Hassell-and-Kernell,-Veto-Rhetoric-and-Legislative-Riders%288-27-2013%29.pdf (discussing veto threat).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
99
The role of the Senate in appointments provides an additional point of
leverage for Congress over agencies.269 Senators can use their power over
these appointments to influence the candidates who are nominated.
Especially during periods of divided government, the Senate confirmation
process has the potential to increase the ideological distance between the
President and Senate-confirmed political appointees. Although empirical
confirmation of the effect of the Senate on administrative agency
appointments is extremely difficult, efforts have been made to examine the
influence of the Senate on judicial appointments.270 This literature has
identified a statistical connection between political preferences in the
Senate at the time of a judge’s nomination and case outcomes for that
judge. 271 The Senate confirmation process can also be used to prod the
administration on unrelated policy matters.272
Finally, Congress’s hearing and investigative powers are another
source of influence. There is no congressional equivalent to the President’s
bully pulpit, but there are opportunities for Congress to focus public
scrutiny on administrative agencies.273 Although “mass media attention
does not come easily” for oversight activities, there are many high-profile
examples in which committees have successfully used hearings to raise the
political salience of an issue.274
269. Nolan McCarty & Rose Razaghian, Advice and Consent: Senate Responses to Executive
Branch Nominations, 1885–1996, 43 AM. J. POL. SCI. 1122, 1124 (1999).
270. See, e.g., Micheal W. Giles, Virginia A. Hettinger & Todd Peppers, Picking Federal
Judges: A Note on Policy and Partisan Selection Agendas, 54 POL. RES. Q. 623 (2001).
271. See Donald R. Songer & Martha Humphries Ginn, Assessing the Impact of Presidential and
Home State Influences on Judicial Decisionmaking in the United States Courts of Appeals, 55 POL.
RES. Q. 299 (2002). But see Joshua B. Fischman & David S. Law, What Is Judicial Ideology, and How
Should We Measure It?, 29 WASH. U. J.L. & POL’Y 133, 196–97 (2009) (finding relatively modest
predictive gains from using “common space scores” based in part on political preferences in the Senate
over the party of the appointing president).
272. For example, in recent years, Democratic senators stalled President Obama’s appointments
to protest administration policy on offshore oil drilling and relationships with Cuba. Lewis, supra note
58, at 587–88 (Mary Landrieu on oil drilling; Robert Menendez on Cuba). Professor and former OIRA
Administrator Cass Sunstein describes his experience during the nomination process as beginning with
a “nightmarish process known as vetting” and culminating in a series of holds by Republican senators
meant to extract concessions from the administration. CASS R. SUNSTEIN, SIMPLER: THE FUTURE OF
GOVERNMENT 19–26 (2013).
273. In addition to reaching the public, hearings can be used to communicate congressional
intentions to the agency or to communicate committee views to the rest of Congress. See Charles M.
Cameron & B. Peter Rosendorff, A Signaling Theory of Congressional Oversight, 5 GAMES & ECON.
BEHAV. 44 (1993). Empirical analysis suggests that these hearings have an effect. See Mary K. Olson,
Agency Rulemaking, Political Influences, Regulation, and Industry Compliance, 15 J.L. ECON. & ORG.
573 (1999) (increased oversight led to heightened enforcement and greater compliance); Jeffery C.
Talbert, Bryan D. Jones & Frank R. Baumgartner, Nonlegislative Hearings and Policy Change in
Congress, 39 AM. J. POL. SCI. 383 (1995); Brian D. Feinstein, Congressional Control of Administrative
Agencies (2014) (unpublished manuscript) (on file with the Alabama Law Review).
274. ABERBACH, supra note 259, at 119.
2 LIVERMORE 45-134 (DO NOT DELETE)
100
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
Congress also takes steps to build on these constitutional powers.
Internal reorganizations have been repeatedly undertaken to increase
Congress’s power free from judicial scrutiny. Most important is the
committee structure, which began at the time of the founding and has gone
through successive waves of restructuring.275 The Legislative
Reorganization Act of 1946, which was explicitly undertaken to enhance
legislative oversight, was adopted the same year as the Administrative
Procedure Act. Congress has also built out its internal bureaucracy through
increased budgets for committee and member staff, and through the
creation and funding of entities such as the Congressional Budget Office
and the Government Accountability Office. The Congressional Review
Act, adopted in the wake of INS v. Chadha, is another attempt (albeit less
successful) at enhancing review over agency actions.
B. Party Discipline in Congress
The oversight tools discussed above are exercised by a diverse body,
not a unitary actor.276 The paradigmatic legislative action is passage of a
bill, along with the associated procedures of committee and floor votes.
Oversight activity is inherently more decentralized. Only certain oversight
activities, such as votes on appropriations or confirmation (in the Senate),
require action on the floor. Other important oversight is carried out on the
committee or subcommittee level, and individual members can pursue
certain oversight activities by themselves. For this reason, the concept of
congressional influence over agencies is hard to pin down. Agency
responsiveness to congressional majorities, congressional leadership,
committee leadership, or individual members might all be understood as
“congressional” influence. Institutional arrangements, including the
internal structure of Congress as well as the organization of political
parties, can affect both the total amount of congressional influence (broadly
understood) on administrative agencies and how that influence is
distributed within the body.277
275. The committee structure enhances control through specialization, but also results in
decentralization, which, given member career incentives, can end up reducing effectiveness. Lawrence
C. Dodd, Congress and the Quest for Power, in CONGRESS RECONSIDERED, supra note 118, at 269,
277–83.
276. ABERBACH, supra note 259, at 23 (noting conflict between “individual versus collective
interests” in Congress).
277. In general, power in Congress can be distributed along roughly three lines: according to the
“institutional median,” according to “allocation politics” based on the strength of member preferences,
and according to the “partisan median” of the majority party. See Doo-Rae Kim, Political Control and
Bureaucratic Autonomy Revisited: A Multi-Institutional Analysis of OSHA Enforcement, 18 J. PUB.
ADMIN. RES. & THEORY 33 (2008).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
101
Irrespective of how power is distributed within Congress, the
constitutional structure of shared oversight forces the President to jockey
for position with various legislative actors in an ongoing struggle over the
reins of the bureaucracy. At various points in U.S. history, Congress, its
committees, and its members have decidedly had the upper hand vis-à-vis
the presidency. For example, during the late nineteenth century, Congress
was able to assert such strong control over the bureaucracy that Woodrow
Wilson—in his pre-politician days as a scholar of American political
institutions—lamented that “congressional government” led to a host of
pathologies that could only be addressed through reforms leading to a more
parliament-like system.278
In subsequent decades, generations of political scientists have followed
Wilson in studying interactions between the President, Congress, and the
bureaucracy.279 A particularly influential group of scholars writing in the
1970s and 1980s developed a set of empirical and normative observations
concerning Congress, administrative agencies, and the President that
challenged dominant understandings of how these institutions operated and
interacted.280 This group of scholars in economics, political science, and
law looked to the behavior of the institutions and actors around them. And
just as Woodrow Wilson’s views were influenced by the particular
circumstances of his day, this group was affected by the organization of
political life during the decades when their ideas were formed.
Scholars during this period studied the federal government at a time
when traditional political parties were in a severe state of decline and
contemporary political parties were just beginning to take form. What they
saw when they studied the institutions around them was constituent- (rather
than party-) oriented members of Congress, powerful seniority- (rather than
party-) based congressional committees, captured or aggrandizing
bureaucrats, weak or absentee presidents, a disorganized public, and
interest-group dominance.281 Together, their findings painted a highly
278. See WILSON, supra note 260.
279. See, e.g., Ferejohn & Shipan, supra note 262; Mathew D. McCubbins & Thomas Schwartz,
Congressional Oversight Overlooked: Police Patrols Versus Fire Alarms, 28 AM. J. POL. SCI. 165
(1984); Craig Volden, A Formal Model of the Politics of Delegation in a Separation of Powers System,
46 AM. J. POL. SCI. 111 (2002); William F. West & Joseph Cooper, Legislative Influence v. Presidential
Dominance: Competing Models of Bureaucratic Control, 104 POL. SCI. Q. 581 (1989).
280. Particularly influential were rational choice accounts that modeled legislators and regulators
as rational, self-promoting actors in political markets in which regulation is traded. See Sam Peltzman,
Toward a More General Theory of Regulation, 19 J.L. & ECON. 211 (1976); Richard A. Posner,
Theories of Economic Regulation, 5 BELL J. ECON. & MGMT. SCI. 335 (1974); George J. Stigler, The
Theory of Economic Regulation, 2 BELL J. ECON. & MGMT. SCI. 3 (1971).
281. See THEODORE J. LOWI, THE END OF LIBERALISM 55–93 (1969) (discussing interest group
domination); ELECTORAL CONNECTION, supra note 95 (arguing that members of Congress hewed
closely to the desires of their constituents); WILLIAM A. NISKANEN, JR., BUREAUCRACY AND
2 LIVERMORE 45-134 (DO NOT DELETE)
102
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
evocative picture of an “iron triangle” of influence-trading between
agencies, the industries they regulated, and the congressional committees
that oversaw them.282 This iron triangle was criticized for benefiting the
powerful and well-organized at the expense of broad public interests such
as consumers and workers.283
A new generation of scholars has revisited the claims that political
parties play very little role in structuring legislative politics, especially in
light of the rise of contemporary parties.284 Perhaps the two most influential
accounts of parties in Congress in recent years are Cox and McCubbins’s
“cartel” theory and Rohde’s conditional party government model.285 Under
the cartel model, legislative parties are a mechanism for members of
Congress to solve collective action problems through “negative agenda
control.”286 Legislative offices with “special agenda-setting powers”—such
as the Speaker and important committee chairmanships—are selected by
the majority party, and the occupants exercise delegated authority on behalf
REPRESENTATIVE GOVERNMENT 53 (1971) (budget-maximizing model of government bureaucrats);
MANCUR OLSON, THE LOGIC OF COLLECTIVE ACTION: PUBLIC GOODS AND THE THEORY OF GROUPS
(1965) (on the overrepresentation of small, concentrated interests vis-à-vis diffuse interests); Kenneth
A. Shepsle & Barry Weingast, Legislative Politics and Budget Outcomes, in FEDERAL BUDGET POLICY
IN THE 1980S 343, 351 (Gregory B. Mills & John L. Palmer eds., 1984) (on committee power); Lance T.
Leloup & Steven A. Shull, Congress Versus the Executive: The “Two Presidencies” Reconsidered, 59
SOC. SCI. Q. 704 (1979) (suggesting that the President is often hamstrung in carrying out a domestic
policy agenda). The relative autonomy of agencies was a particular source of interest. See generally
ROBERT A. KATZMANN, REGULATORY BUREAUCRACY (1980); James Q. Wilson, The Rise of the
Bureaucratic State, 41 PUB. INT. 77 (1975).
282. This three-sided relationship is sometimes referred to in the political science literature as
“subgovernments.” See Daniel McCool, Subgovernments as Determinants for Political Viability, 105
POL. SCI. Q. 269, 269 (1990); see also Andrew S. McFarland, Interest Groups and Theories of Power in
America, 17 BRIT. J. POL. SCI. 129, 134 (1987) (providing a more general account of interest group
politics that includes “plural elitism” as one of several possible manifestations). In general, this group of
scholars rejected both benign pluralism and public-spirited social planners as accurate representations
of Congress and the bureaucracy. See generally EDWARD C. BANFIELD, POLITICAL INFLUENCE (1961);
DAVID BRAYBROOKE & CHARLES E. LINDBLOM, A STRATEGY OF DECISION (1963); ROBERT A. DAHL,
A PREFACE TO DEMOCRATIC THEORY (1956); ROBERT A. DAHL, WHO GOVERNS? (1961); NELSON W.
POLSBY, CONGRESS AND THE PRESIDENCY (3d ed.1976); DAVID B. TRUMAN, THE GOVERNMENTAL
PROCESS (2d ed.1971); AARON B. WILDAVSKY, THE POLITICS OF THE BUDGETARY PROCESS (1964).
283. See Stewart, supra note 68.
284. See, e.g., D. RODERICK KIEWIET & MATHEW D. MCCUBBINS, THE LOGIC OF DELEGATION:
CONGRESSIONAL PARTIES AND THE APPROPRIATIONS PROCESS 22–38 (1991) (arguing that delegation
can forward the policy goals of the majority party in Congress, undercutting the claim that delegation
implies “abdication” on the part of the legislature). Not every scholar writing in recent decades has
come to the conclusion that parties are the main driving force of congressional organization. See KEITH
KREHBIEL, INFORMATION AND LEGISLATIVE ORGANIZATION (1991) (pursing an informational account
of legislative organization in which structure forwards the preferences of the body as a whole, rather
than a partisan coalition).
285. See LEGISLATIVE LEVIATHAN, supra note 94, at 2; SETTING THE AGENDA, supra note 94, at
38; DAVID W. ROHDE, PARTIES AND LEADERS IN THE POSTREFORM HOUSE 31 (1991).
286. SETTING THE AGENDA, supra note 94, at 101.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
103
of the party.287 Defection from the party agenda is assured not by policing
member voting, but by only allowing legislation that will generate
conforming behavior to reach the floor.288
For Rohde’s conditional party government model, party dominance
only arises when there is sufficiently homogenous party membership
together with an institutional context that places power in the hands of a
leader who is responsive to the median-party member.289 Without sufficient
agreement among members, party coalitions fracture, and the power for a
stable majority coalition to exercise control is latent. Rohde’s account helps
explain periods of both more united, party-dominated Congresses and more
decentralized, committee-dominated Congresses.290 Rohde also produces
evidence that the contemporary period—starting with the transition of the
South to the Republican Party and the House rules reforms in the 1970s—is
one in which the conditions necessary for conditional party government
have been met.291 If this is the case, we can expect those congressional
institutions that are subject to majority control—such as the House
leadership and the committees—to be responsive to the median member of
the controlling party.
If parties do play a role in structuring legislative politics more
generally, the question remains whether and how parties affect oversight
activities. Although some oversight activities require the whole Congress to
act, much of the prosaic oversight work is carried out by committees and
individual members. The mechanisms of negative agenda control
hypothesized in Cox and McCubbins’s cartel theory, for example, would
not function as an effective tool for resisting defection concerning
oversight. Unlike in the context of legislation, where committees serve as
gatekeepers for “floor” activity, individual members can pursue at least
some oversight activities on their own initiative. Although certain
activities—such as votes on presidential nominees, appropriations riders,
and legislation that affects an agency’s operations—must pass through
committee gates, other oversight is much more decentralized and is
frequently undertaken at the subcommittee and individual levels.292
287. Id. at 24–25.
288. Id. (“[C]artel members expect those appointed to agenda-setting offices to always obey ‘the
first commandment of party leadership’ – Thou shalt not aid bills that will split thy party – and to
sometimes obey the second commandment – Thou shalt aid bills that most in thy party like.”).
289. ROHDE, supra note 285, at 169.
290. Id.
291. Id. For information on the congressional reforms of the 1970s, see Eric Schickler, Eric
McGhee & John Sides, Remaking the House and Senate: Personal Power, Ideology, and the 1970s
Reforms, 28 LEGIS. STUD. Q. 297 (2003).
292. See generally ABERBACH, supra note 259 (explaining common oversight activities, few of
which involve floor votes).
2 LIVERMORE 45-134 (DO NOT DELETE)
104
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
Rhode’s theory of conditional party government, where homogeneity
among members is a causal factor, is more consistent with a stronger role
for partisan considerations in oversight. Where co-partisans share a moreor-less homogenous set of preferences, even their decentralized oversight
activities will conform to partisan expectations.293
The internal organization of Congress also likely affects the degree to
which partisan considerations affect oversight activities, compared to more
particularistic, constituent-driven concerns. In this respect, the reforms
ushered in during the 94th Congress, which included greater authority by
legislative leadership over the appointment of committee chairs, a shifting
of expertise away from oversight committees, and the reduction of
committee staff, may have been particularly important in harmonizing
oversight activities with the broader party program.294
C. Interbranch Sparring
The recent emergence of a more obviously partisan dynamic in
Washington, D.C., has increased attention on the role of political parties
and their place in the constitutional order.295 Some have expressed worries
that partisanship will exacerbate conflict built into the separation-of-powers
constitutional structure and inhibit government effectiveness.296 For
example, in 2001, Kagan argued that partisan motivations can be
“superimposed on institutional differences” and exacerbate interbranch
conflict during periods of divided government.297 Others have argued that
293. It bears noting that co-partisans share incentives to coordinate their behavior on even many
non-ideologically charged issues where coordination enhances their standing vis-à-vis the opposition.
See FRANCES E. LEE, BEYOND IDEOLOGY: POLITICS, PRINCIPLES, AND PARTISANSHIP IN THE U.S.
SENATE 133 (2009).
294. See generally THOMAS E. MANN & NORMAN J. ORNSTEIN, THE BROKEN BRANCH: HOW
CONGRESS IS FAILING AMERICA AND HOW TO GET IT BACK ON TRACK (2006).
295. See, e.g., Jack M. Balkin, The Last Days of Disco: Why the American Political System Is
Dysfunctional, 94 B.U. L. REV. 1159 (2014). Professor Bulman-Pozen has recently argued that
polarization has allowed partisan differences to be transposed onto the federalist structure “in ways that
articulate, stage, and amplify competition between the political parties” in generally beneficial ways.
Bulman-Pozen, supra note 126, at 1080.
296. See Peter L. Strauss, The Place of Agencies in Government: Separation of Powers and the
Fourth Branch, 84 COLUM. L. REV. 573, 664–65 (1984).
297. Kagan, supra note 3, at 2344. For Professor Mark Tushnet, the contemporary period of
divided government is marked by “subdued constitutional ambition” in which “the aspiration of
achieving justice directly through law has been substantially chastened.” Mark Tushnet, Foreword: The
New Constitutional Order and the Chastening of Constitutional Aspiration, 113 HARV. L. REV. 29, 33
(1999). Tushnet extends this argument in his subsequent book, THE NEW CONSTITUTIONAL ORDER
(2003).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
105
partisanship suppresses essential competition between the branches during
periods of united government.298
A lively empirical debate, initiated by Professor Mayhew in 1991,
seeks to determine whether either of these fears is grounded. 299 Mayhew
finds that the federal government functions similarly under united and
divided regimes, with equivalent levels of oversight and legislative
productivity.300 Subsequent research has challenged those findings with
evidence that partisan composition does indeed affect legislative behavior,
with greater production of statutes during periods of united government and
more rigorous oversight during periods of divided government.301 Recent
history provides stark support for the latter view. During President
Obama’s first two years in office, when Democrats enjoyed substantial
majorities in both legislative chambers, major, sweeping pieces of
legislation, including the Affordable Care Act and the Dodd-Frank
financial reforms, were adopted. After the Republican Party gained a
majority in the House in 2010, statutory production fell to essentially nil,
and interbranch hostility rose considerably.
Focusing on administrative law, Professor McGarity contrasts a
“deliberative, lawyer-dominated domain of traditional administrative law”
with “blood-sport confrontations over agency rulemaking.”302 The former
298. See Daryl J. Levinson & Richard H. Pildes, Separation of Parties, Not Powers, 119 HARV.
L. REV. 2312 (2006); see also Douglas Kriner, Can Enhanced Oversight Repair “The Broken
Branch”?, 89 B. U. L. REV. 765, 783 (2009) (focusing on lack of oversight of the Iraq War during the
united Republican government). The inverse is sometimes argued. According to the “responsible
government model,” which looks to parliamentary systems of government for inspiration, unified
government, by harmonizing incentives between the majority party in Congress and the President, is
thought to overcome the pathological tendencies of overly antagonistic branches.
299. See DAVID R. MAYHEW, DIVIDED WE GOVERN: PARTY CONTROL, LAWMAKING, AND
INVESTIGATIONS, 1946–2002 178 (2d ed. 2005) (finding that periods of divided government are not less
productive).
300. Id.
301.
See John J. Coleman, Unified Government, Divided Government, and Party
Responsiveness, 93 AM. POL. SCI. REV. 821, 832–33 (1999) (finding greater legislative productivity and
responsiveness to public opinion during period of united government); Sean Q. Kelly, Divided We
Govern? A Reassessment, 25 POLITY 475, 482–83 (1993) (reexamination of Mayhew’s data finding
significant differences between divided and united governments); Douglas Kriner & Liam Schwartz,
Divided Government and Congressional Investigations, 33 LEGIS. STUD. Q. 295, 314 (2008) (finding
that investigations respond to both the partisan composition and the cohesiveness of the majority party).
For an early account, see Seymour Scher, Conditions for Legislative Control, 25 J. POL. 526, 537–38
(1963) (noting that the 1960 election of John F. Kennedy substantially decreased incentives for the
Democratic House to scrutinize agency action). See also Neal Devins, Party Polarization and
Congressional Committee Consideration of Constitutional Questions, 105 NW. U. L. REV. 737 (2011).
302. Thomas O. McGarity, Administrative Law as Blood Sport: Policy Erosion in a Highly
Partisan Age, 61 DUKE L.J. 1671, 1680, 1711–12 (2012); see also Kagan, supra note 3, at 2346–49
(arguing that during periods of divided government, members of Congress face partisan incentives to
focus oversight activities on regulatory matters that have been signaled to be high priorities for the
President).
2 LIVERMORE 45-134 (DO NOT DELETE)
106
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
approach manifested during the period when political parties played a less
dominant role in congressional oversight.303 As parties have grown in
prominence, the “blood sport” has commenced.304 As should be clear from
the characterization of the contemporary period, McGarity finds it
substantially less to his taste than the bygone era of traditional
administrative law. In a recent essay, Professor Cass Sunstein sounds
similar notes, coining the word “partyism” to describe the current level of
partisanship in the nation’s culture and comparing partyism with other
“isms” like sexism and racism—essentially a bias that undermines
deliberation and rational thought.305
The blood-sport and partyism metaphors focus on the potential vices of
political parties for the relationship between presidential and congressional
oversight of agencies, but they do not present a general normative
framework for evaluating how Congress interacts with agencies.
McGarity’s “deliberative, lawyer-dominated domain” is likely unattainable,
given the reality of contemporary political parties and the incentives for
congressional actors to challenge the actions of the sitting President. If that
is the only normative model, then the best that can be done is to limit
congressional influence to the absolute minimum. But an alternative vision
that focuses on the virtues of parties is also possible. This alternative is
related to the potential normative justification given in Part III.C for
presidential administration that is grounded in the concept of responsible
party government.
With respect to presidential oversight, responsible party government is
facilitated by a relatively free rein given to presidents to control
administrative agencies. It is this free rein that makes it possible for a
political party, after gaining office in a fair election, to implement its
program. The link between elections and outcomes is what allows
responsible party government to exist.306
Congress–President relations can also be examined through a
responsible party government lens. Under this model, party competition
generates alternative regulatory programs that are tested for popular
support during elections, and while the winner implements the program, the
losing party takes up positions in Congress to criticize the President and
revise its program to improve its electoral appeal. Members of Congress in
the out-of-presidency party use the platform given to them by their
legislative roles to act in a similar vein to the shadow cabinet familiar from
303.
304.
305.
306.
McGarity, supra note 302, at 1680.
Id.
Sunstein, supra note 28.
Cf. SCHATTSCHNEIDER, supra note 247, at 69.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
107
parliamentary systems such as the United Kingdom. The role of the shadow
cabinet is perhaps most vividly manifested in the tradition of “question
time,” with its origins in the Westminster system, in which the Prime
Minister and other government officials submit to periodic questioning by
members of Parliament from the opposition party.307 The shadow cabinet
lacks a substantive role in government, but it enhances the public
accountability of the majority party through constant scrutiny and criticism.
Although the responsible party government model highlights the
democratic potential in a system of party-motivated opposition to the
President’s regulatory program, adoption of a new normative model does
not settle the matter. Rather, it reframes the inquiry. The question becomes
whether current arrangements deliver on the promise of responsible party
government, and whether those benefits justify the costs that are generated
along the way.
In making that calculation, perhaps the most important consideration is
that, in the U.S. constitutional system, an out-of-presidency party still
enjoys substantial power. In a parliamentary system, the majority party
holds all of the reins of power—the opposition has a voice, but little else.
Not so in the United States, where even a party in the minority in both
legislative houses can still use super-majority rules and member
prerogative to exert influence. If an out-of-White-House party gains a
majority in either legislative body, its power increases substantially. If
opposition parties consistently find it to their advantage to obstruct
altogether the actions of the executive, rather than negotiate or criticize, it
has the potential to severely undermine the effectiveness of the
government.308
The zenith of this obstructionist risk would occur if a stable
equilibrium of divided control came about in which one of the major
parties adopted a program of protecting the status quo against reform. Such
a status-quo-protecting party would wield substantial power, given the U.S.
constitutional structure’s inherent bias toward inaction. And since merely
blocking changes to the status quo would not require coordination with the
other party, a simple majority (or even sub-majority) in either chamber
could be translated into a veto wielded with little need for compromise.
307. For a history of the practice of parliamentary questions, which have been in place in their
modern form in the U.K. since the early nineteenth century, see D. N. CHESTER & NONA BOWRING,
QUESTIONS IN PARLIAMENT (1962).
308. Research on the U.S. Senate has found that there do appear to be partisan motivations on the
part of out-of-White-House senators to oppose the President’s program. These partisan motivations
exist even on issues of widespread policy agreement; the mere fact that the President has championed
an issue induces opposite-party senators to take an oppositional stance. See generally LEE, supra note
293 (undertaking extensive evaluation of behavior in the U.S. Senate). This fact does not bode well for
civil and productive oversight.
2 LIVERMORE 45-134 (DO NOT DELETE)
108
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
The models of legislative party discipline discussed in the previous
section also bear on whether the partisan oversight dynamic between the
President and Congress is better described as blood sport or loyal
opposition. There are, generally speaking, three sets of preferences that
members must balance. The first includes those of the median voter in their
districts; responsiveness to these preferences is predicted by Mayhew’s
electoral connection model of legislative politics.309 The second includes
those of party actors operating through the primary or caucus system.310
The third is the preferences of the median co-partisan in Congress; this is
predicted by Cox and McCubbins’s cartel theory and Rohde’s conditional
party government theory.311
A responsible party government approach to congressional oversight
seems most likely to arise if members are responsive to the party median in
the legislature, as opposed to alternative sources of pressure.
Responsiveness to constituent medians is likely to devolve into the kind of
particularistic claims on the administrative state that drove earlier criticisms
of congressional oversight.312 The aggregate collection of particularistic
claims, likely closely related to regional concerns rather than programmatic
agendas, do not add up to the comprehensive program that is consistent
with responsible party government.313
Responsiveness to discipline from external, non-legislative party
actors, on the other hand, runs a greater risk of escalating into the blood
sport and partyism that McGarity and Sunstein fear. Schlesinger describes
centrifugal and centripetal pressures within political parties that,
respectively, pull parties away from and toward the median voters in the
national constituency.314 Centrifugal force is generated by the desire of
politicians to be elected to office and drives parties to adopt programs that
are attractive to a broad swath of the electorate. Centripetal force is
generated by the activist base of a party: the donors, volunteers, and policy
networks that pull party programs toward their outlier preferences.
Schlesinger believes that centripetal forces can be important, but that in the
final analysis, politicians and their desire to remain in office will carry the
309. See PLACING PARTIES, supra note 95.
310. See supra Part I.C.
311. Legislators have their own policy preferences as well, a point stressed by ALDRICH, supra
note 6. In Aldrich’s model, these preferences tend to pull politicians away from median voters and
toward the partisan median—for purposes of the analysis above, they can be treated as exerting a
similar force as intense policy demanders.
312. See Kagan, supra note 3.
313. See JOSEPH P. HARRIS, CONGRESSIONAL CONTROL OF ADMINISTRATION (1964) (arguing
that Congress pursues narrow interests and that Congress can hamper executive action that is more
nationally oriented).
314. See SCHLESINGER, supra note 217, at 177–79.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
109
day on most issues.315 But Schlesinger’s faith may be misplaced or
outdated; if contemporary political parties have become so dominated by
intense policy demanders that they systematically sacrifice the broad
national interest in smooth governance to engage in obstructionism or
“message politics,”316 concerns about an era of blood-sport politics may be
well founded.
If the conditions exist for politicians—and especially centrist voices
within each party—to resist particularistic and centrifugal pressures
through party structures, the risks of a blood sport are reduced. The
question is whether contemporary political parties give politicians the tools
they need to resist the two sets of pressures that inhibit responsible party
government. There are good reasons to believe that particularistic pressures
are less important than in the past; by many measures, politicians in both
parties conform to their party programs, even at the expense of local
benefits. At the same time, the centrifugal pressure from intense policy
demanders has increased, and high-profile examples of party leaders losing
their seats to primary challengers serve as a reminder that the balance of
power within parties between politicians and intense policy demanders
does not tilt definitively toward politicians.317
Just as the structure of political parties affects the ability of the
President to exercise managerial control over agencies and the justification
for presidential control, it also affects relationships with Congress. A
Congress that is run by party-organized and electorally oriented legislators
can be expected to engage in relatively more productive critical sparring
with the President over the policy course that is set in a given
administration. When in the minority, the out-of-presidency party will
nevertheless be able to give voice to criticisms of the President’s program
and articulate an alternative vision. When in the majority, the out-ofpresidency party will use its power to increase the volume of its criticism,
and, at least part of the time, to engage in productive dialogue and
compromise to expand its electoral appeal. All of these behaviors are
highly consistent with a responsible party government model. Then again,
if Congress is run by politicians who are in thrall to or cowed by intense
policy demanders—especially those intent on status-quo-protecting
obstructionism or symbolic message politics divorced from policy
outcomes—then congressional oversight is more likely to resemble a blood
sport than productive policy disagreement.
315. Id.
316. Devins, supra note 301, at 758.
317. See generally Richard H. Pidles, Romanticizing Democracy, Political Fragmentation, and
the Decline of American Government, 124 YALE L.J. 804 (2014) (discussing mechanisms to enhance
centralized party control and avoid outside pressures on politicians to tilt toward extreme views).
2 LIVERMORE 45-134 (DO NOT DELETE)
110
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
V. RESPONSIBLE PARTY ADMINISTRATION
For some time, commentators have recognized deep tensions at the
heart of the administrative state.318 Agencies are asked, on the one hand, to
be democratically responsive, and on the other, to carry out their
responsibilities with neutral expertise. They should maintain fidelity to the
law but also administer fragmented statutory authority with vigor and
coherence. Under the most optimistic versions of presidential leadership,
White House oversight mitigates these tensions, but the reality of
contemporary parties undermines that happy narrative.319 Traditional
administrative law provides an alternative approach, grounded in
procedural and substantive norms that protect legality, public participation,
and neutrality. But, while this approach has strong attraction, it has not
adapted to new realities, including the centrality of contemporary parties in
the administrative state.320
At the same time, although responsible party government may help
define normative aspirations for contemporary political parties, it is
incomplete (at best) as a model of administration. The oversight roles of the
President and Congress create opportunities for parties to develop and
implement programmatic agendas while in control of the White House and
criticize the other party when out of power. These oversight roles help
facilitate responsible party government in a constitutional system in which
power is fractured between the legislative and executive branches. But
agencies are typically thought to be more than instruments to implement
party programs—other values central to the legitimacy of the
administrative state are incompletely captured by, and sometimes
contradict with, responsiveness to party programs.
This Part proposes and applies a normative account of administrative
law that seeks to integrate party government with traditional administrative
values. This responsible party administration framework evaluates legal
arrangements based on how well they facilitate responsible party
government while protecting important administrative values, such as
expertise. This approach can be applied to a host of questions both within
and outside of the area of administrative law, from institutional design to
legal doctrine. Two particularly important sets of issues that affect
responsible party administration are doctrines concerning judicial deference
318. Herbert Kaufman, Emerging Conflicts in the Doctrines of Public Administration, 50 AM.
POL. SCI. REV. 1057, 1057 (1956).
319. See supra Part III.
320. See generally Farber & O’Connell, supra note 5.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
111
and the practice of regulatory review, both of which are treated in some
detail below.
A. Administrative Values and Party Rivalry
Six decades ago, Professor Herbert Kaufman described “three core
values” that structure the organization and operation of administrative
institutions in the United States. He designated those values as
“representativeness, neutral competence, and executive leadership.”321 The
final value may be best understood as coherence and vigor—principles that
might be brought about by strong executive leadership.322 In a recent
survey, Professor Vermeule offers “constitutionality,” “democratic
credentials,” and “epistemic capacity” as the dominant normative
frameworks for understanding the administrative state.323 The principles
stressed by Kaufman and Vermeule can be thought of as clusters of
normative goals that capture a broad spectrum of aspirations and
commitments that have motivated analysis, justification, and reform of
agencies and oversight institutions for many years.324 For purposes of
analysis, the following discussion draws four categories of administrative
values: coherence and vigor, promoted by executive leadership;
representativeness, promoted by some form of democratic accountability;
expertise, promoted by professionalism and substantive rigor; and legality,
which requires constitutional and statutory pedigree for administrative
action.
321. Kaufman, supra note 318.
322. Id.
323. Adrian Vermeule, The Administrative State: Law, Democracy, and Knowledge, in THE
OXFORD HANDBOOK OF THE U.S. CONSTITUTION (Mark Tushnet, Sanford Levinson, & Mark A.
Graber, eds. 2015). The overlap of the two scholars, writing several decades apart, is instructive: the
complex interplay of democratic accountability and expertise is an enduring concern for the
administrative state. The divergence is instructive as well—Kaufman does not carve out a distinct
category of constitutionality as an administrative value, which may reflect a period of relative
consensus concerning the desirability of strong national administrative institutions. Vermeule, writing
during a period of presidential ascendency, does not stress coherence and vigor, which were dominant
considerations in earlier years.
324. Other normative perspectives, such as welfarism and civic republicanism, can and have
been used to evaluate institutional design in the administrative state. See, e.g., MATTHEW D. ADLER &
ERIC. A. POSNER, NEW FOUNDATIONS OF COST-BENEFIT ANALYSIS (2006) (welfarism); Mark B.
Seidenfeld, A Civic Republican Justification for the Bureaucratic State, 105 HARV. L. REV. 1511
(1992) (civic republicanism). These normative frameworks could be, to some extent, captured by the
Kaufman/Vermuele values—for example, representativeness could be thought to be related to certain
interpretations of welfarism where actual preferences serve as the basis for welfare determinations. Or,
alternatively these institutional values might be thought to be justified themselves if they facilitate
welfare maximizing outcomes. In any case, the four values are used here to represent classic
administrative law considerations, not the whole scope of potential normative theories that could be
brought to bear on the administrative state. Cf. Vermuele, supra note 323.
2 LIVERMORE 45-134 (DO NOT DELETE)
112
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
Administrative values sometimes coexist comfortably and sometimes
conflict. Presidential power and neutral competence were a comfortable fit
during the Progressive Era, when measures to dismantle the spoils system
(such as civil service employment protections) simultaneously shored up
professionalism within the bureaucracy and sapped authority from
traditional party organizations that competed for power with the President.
At the same time, it is hard to deny the potential for conflict between
administrative law values.325 Democratic responsiveness and neutral
expertise, for example, are not obvious bedfellows. 326
Contemporary political parties have ambiguous effects on these
administrative law values and the tension between them. First,
contemporary parties both enhance and interfere with the values of
coherence and vigor promoted by executive leadership. Partisans of more
programmatic parties may use their positions in Congress to thwart the
President’s agenda, sap resources away from administrative agencies,
embarrass the White House, starve agencies of necessary resources, deny
statutory authorization, hold up the federal budget, impede essential debt
financing, or even undermine foreign policy initiatives. Collectively, these
efforts drain the ability of presidents to exercise leadership and reduce
presidential and agency aspirations. At the same time, as discussed in Part
II, contemporary parties boost presidential power over agencies—placing
hiring decisions more firmly under White House control, providing a bank
of loyal and competent personnel, and generating innovative programmatic
platforms for presidential administrations to pursue while in power.
The second administrative value, representativeness, is also placed
under multiple countervailing pressures by contemporary parties.
Programmatic parties may foster representativeness by clarifying lines of
accountability and establishing the regulatory policy stakes of elections. At
the same time, the structure of contemporary parties may pull presidential
and congressional preferences away from the center and toward the desires
325. For its most ardent defenders, presidential administration manages to overcome all of the
potential tensions between administrative values simultaneously. Under this account, discussed in detail
in Part III, presidents bring a coherent vision and vitality to agencies, in part because they must be
responsive to a broad national constituency. At the same time, presidents have incentives to promote
neutral competence in administration to deliver on their promises to the electorate. To this union of
Kaufman’s three values (which is at the heart of Justice Kagan’s defense of presidential administration,
cf. Kagan, supra note 3) we might add unitary executive constitutional interpretations that attempt to
reconcile the vast contemporary federal bureaucracy with the constitutional structure.
326. During his time, Kaufman noted an “emerging conflict” between executive leadership and
neutral competence because presidents, “in whom administrative responsibility and power were to be
lodged, were also partisan politicians.” Kaufman, supra note 318, at 1067. Kaufman predicted that,
“[p]olitical scientists of the . . . future, looking back, may well conclude that it is not easy to bridge the
gap between . . . seeking to encourage the growth of a professional bureaucracy and . . . turmoil over
how to control it.” Id. at 1073.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
113
of party activists and intense policy demanders. To the extent that the
extreme donor and volunteer bases of the two political parties dominate
more moderate voices, political control over agencies will tend to pull
administrative decision making away from majority preferences.
Contemporary parties have similarly ambiguous effects on neutral
competence. As discussed in Part II, contemporary parties comprise some
entities—such as advocacy organizations and think tanks—that are deeply
embedded within the issue networks associated with particular
administrative agencies (or groups of agencies). These issue-network
entities cultivate policy expertise in the arcane and highly technical matters
that are at the center of many administrative proceedings. This technical
expertise helps ensure that political appointees are not mere “amateurs”
with no choice but to defer to civil service experts purportedly under their
control.327 Instead, they come into office with the ability to competently
manage and evaluate the performance of the departments they supervise.
But this competence is not neutral; it is applied to forward programmatic
goals associated with the competing parties. Technocratic competence
provides political appointees with the means to effectively oversee
agencies, but party programs provide the ends to which that oversight
authority is used. Furthermore, as political appointees penetrate deeper into
administrative agencies, they displace and undermine the career incentives
of permanent personnel. To the extent that political points can be scored by
“bureaucrat bashing” and reducing the professional staff lines at agencies,
parties will also have incentives that are antagonistic to agency expertise.328
Finally, from the perspective of legality, contemporary parties play an
even more problematic role. There is no obvious sense in which party
government enhances the legal pedigree of agency action. At the same
time, the strong interest group affiliations and ideological commitments
associated with contemporary parties may undermine the felt obligation on
the part of political appointees to neutrally carry out laws that they disagree
with or that harm favored party constituencies. Similarly, for those with
constitutional concerns about the administrative state, the explicit
incorporation of regulatory policy into party programs may increase the
perception that agency leaders claim “the power to decide . . . which policy
goals [they] wish[] to pursue,” a power critics believe can only be
legitimately exercised by Congress.329
327. See Kaufman, supra note 318, at 1062.
328. See generally R. Sam Garrett, James A. Thurber, A. Lee Fritschler & David H.
Rosenbloom, Assessing the Impact of Bureaucracy Bashing by Electoral Campaigns, 66 PUB. ADMIN.
REV. 228 (2006).
329. Michigan v. EPA, 135 S. Ct. 2699, 2713 (2015) (Thomas, J. concurring) (claiming that
Chevron deference violates constitutional separation of powers principles).
2 LIVERMORE 45-134 (DO NOT DELETE)
114
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
Contemporary parties can also increase tensions between
administrative law values. As described in Part III, the structure of
contemporary political parties undermines at least a straightforward,
median voter–based model of presidential accountability, placing a wedge
between the values of vigor and coherence (often thought to be promoted
by presidential influence), on the one hand, and representativeness on the
other. Similarly, among proponents of neutral competence, many argue that
agencies are the repositories of expertise and impartiality and that
presidential control, whatever its virtues in terms of vigor, will tend to
undermine the role of professional judgment in agency decision making.330
There are two general strategies for reconciling the tensions between
contemporary parties and administrative values. The first is to enhance
party responsibility through institutions and doctrine that promote the
development of normatively attractive forms of party rivalry.331 The
alternative of irresponsible party government is a world of powerful
constituencies within parties that have overbearing influence on
presidential and congressional oversight, leading to agency policy that tilts
toward ideological extremes and congressional oversight that swings
between neglect and gridlock. In this world, party control presents few
virtues, and administrative values are systematically sacrificed for very
little useful end.
But even were responsible party government to be fully realized, there
would still be conflict with administrative values. When a statute is at odds
with a President’s party program, carrying out the party agenda will
conflict with the administrative value of legality. The inevitable overriding
of career bureaucrats by political appointees conflicts with the value of
expertise. Oscillating agency policies during periods of electoral transition
undermine coherence and vigorous implementation of the law. Differing
party programs—essential to responsible party government—ensures that
agency policies will stand at some distance from the median voter,
conflicting with the value of representativeness.
The second strategy attempts to mitigate these conflicts through
responsible party administration that defines a legitimate scope for party
government in a context where administrative values are given their
330. See, e.g., Rena Steinzor, The Case for Abolishing Centralized White House Regulatory
Review, 1 MICH. J. ENVTL. & ADMIN. L. 209 (2012); RENA STEINZOR & SIDNEY SHAPIRO, THE
PEOPLE’S AGENTS AND THE BATTLE TO PROTECT THE AMERICAN PUBLIC 12245 (2010); see also
Barkow, supra note 64, at 19 (“The classic explanation for agency independence is the need for expert
decision making.”); Lisa Schultz Bressman & Robert B. Thompson, The Future of Agency
Independence, 63 VAND. L. REV. 599, 612–13 (2010) (discussing expertise-based justifications for
limiting presidential influence).
331. For a normative account of parties that makes explicit attractive and unattractive forms of
party rivalry, see ROSENBLUM, supra note 91.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
115
appropriate due. For example, the adjudication/rulemaking dichotomy
illustrated in Londoner and Bi-Metallic—foundational cases in American
administrative law—create a separation between spheres that remains
useful in a responsible party administration framework.332 Granting
procedural rights to protect the particularistic, backward-looking context of
adjudication from partisan influence does little to harm responsible party
government and provides substantial benefit to core administrative values
associated with expertise and legality. The general, efficiency-oriented and
future directed context of rulemaking is a better candidate for partisan
influence—more closely related to the goals of responsible party
government and less damaging to administrative values.
Responsible party administration encompasses both strategies: first
improving party responsibility; and second vindicating administrative
values while providing legitimate scope for party government. This
formulation is admittedly vague—the concept of party responsibility is
open to interpretation, as is the best balance of administrative values and
party government. A few additional words can partially clarify the picture.
This is a pragmatic project, in the sense that it does not attempt an account
of administrative law for all times, or under optimal conditions, but instead
seeks to respond to the contemporary political environment in a way that
draws out potential benefits while limiting downside risks. It is also,
therefore, temporary and contingent on current circumstances. Although
instability in administrative law is not desirable, some degree of
unpredictability necessarily accompanies a legal system that evolves
alongside the dynamic political ecosystem which it inhabits. 333
The project is also pluralistic, in the sense that it acknowledges
multiple, competing values that must be balanced and accommodated,
rather than reduced to a single overarching dimension. Perhaps it is
possible to fully reconcile a best possible account of parties with a best
332. See Richard J. Pierce, Jr., Separation of Powers and the Limits of Independence, 30 WM. &
MARY L. REV. 365, 366–67 (1989) (noting that Londoner and Bi-Metallic have long established the
distinction between actions that affect individuals or small groups from those that affect a “large
number of people”).
333. Although there is doubtless some value in developing normative models that are
intentionally abstract, failure by actual decision makers to recognize the distance between contemporary
politics and an ideal democratic system would turn administrative law into Pliny’s ostrich. PLINY,
NATURAL HISTORY bk. 10, ch.1 (Harris Rackham trans., 1938) (observing, in is C.E. 77 discussion of
“the Ostrich,” that “their stupidity is … remarkable; for although the rest of their body is so large, they
imagine, when they have thrust their head and neck into a bush, that the whole of the body is
concealed”). The consequences of legal decisions are affected by political realities, and if consequences
matter, the normative aspirations used to evaluate the law must take account of them as well. Cf. Einer
Elhauge, Does Interest Group Theory Justify More Intrusive Judicial Review?, 101 YALE L.J. 31
(1991); Thomas Merrill, Does Public Choice Theory Justify Judicial Activism After All?, 21 HARV. J. L.
& PUB. POL’Y 219 (1997).
2 LIVERMORE 45-134 (DO NOT DELETE)
116
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
possible account of the administrative state, but this project takes the more
modest goal of highlighting conflicts and clarifying tradeoffs between
competing priorities. If none of the values embedded in responsible party
administration are lexically prior to the others, reforms that make
comparatively substantial improvements along some dimensions with few
countervailing negative effects should be adopted.334 Outside these
contexts, where genuine conflict between competing values arises, one’s
favored resolution will frequently depend on intuitions concerning the
relative importance of one or the other values, a set of debates that will not
be joined here.
Finally, this project explicitly recognizes the reality of partisan
influence over agencies, and allows for the desirability of parties and the
potential for beneficial interaction between parties and the administrative
state.335 The role of politics or contested values in informing administrative
decision making has long been recognized.336 The concept of responsible
party administration specifically recognizes the value of political parties for
a functioning democratic society, and examines administrative law through
that lens. Under the responsible party administrative framework, ceteris
paribus, administrative law reforms that facilitate responsible parties are
desirable. Legal arrangements that impose some cost to traditional
administrative law values may even be acceptable, as long as the benefits to
responsible parties are sufficiently great.
A host of institutional design and doctrinal questions are implicated by
the framework of responsible party administration. Because it favors
responsible parties, it implicates electoral law, campaign finance, civic
engagement, and the law and conventions that structure party and
legislative organization.337 Agency structure helps determine the degree of
exposure of agency decision making to different types of partisan influence
and can also have implications for party responsibility as lines of authority
and accountability are clarified or obscured through institutional design.
Structural decisions implicated by responsible party administration include
the number of political appointees within agencies and the number of
334. Cf. Michael A. Livermore, The Meaning of Green Growth, 3 MICH. J. ENVTL. & ADMIN. L.
33, 59–63 (2013).
335. Compare with Mark Seidenfeld, The Role of Politics in the Deliberative State, 81 GEO.
WASH. L. REV. 1397 (2013) (attempting to examine administrative procedures that allow for political
value to inform administrative decisions, but not examining how administrative law interacts with the
role of parties specifically).
336. See, e.g., Stewart, supra note 68.
337. See generally SOLUTIONS TO POLITICAL POLARIZATION IN AMERICA (Nathaniel Persily ed.,
2015) (providing various policy recommendations that roughly track responsible party government
norms). See also, Ian Vandewalker and Daniel I. Weiner, Stronger Parties, Stronger Democracy:
Rethinking Reform (Brennan Center Working Paper, 2015).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
11/5/2015 9:20 AM
Political Parties and Presidential Oversight
117
positions subject to Senate approval; the norms and conventions that
govern the Senate confirmation process; the incorporation of “equalizing”
factors into agency structure; 338 the distribution of litigating and budgeting
authority in the executive; the internal organization of congressional
oversight and the structure of so-called independent agencies;339 and the
allocation of responsibility to state or federal agencies.340
Rather than attempting to survey this entire landscape, the following
sections focus on two illustrative questions related to agency rulemaking.
The first examines judicial review of agency reasoning-giving alongside
(putative) deference to agency legal interpretations. The second examines
centralized executive oversight. These two issues are important in their
own right and are indicative of the tensions involved in responsible party
administration, but they far from exhaust the possible applications of the
framework.
B. Judicial Review
As discussed in Part I.B, in the 1970s the D.C. Circuit expanded
judicial review of agency rulemaking.341 These requirements included both
procedural elements, such as limits on ex-parte communications for
informal rulemaking, as well as “hard look” review of the substance of
agency policy decisions under the arbitrary and capricious standard.342
Commentators have characterized the goal of courts at this time as
attempting to “reinvent[] the administrative process as a perfected political
process” that facilitated pluralistic bargaining between interest groups and
was removed from the corrupting influence and distorting power
imbalances of the electoral system.343
In 1978, the Supreme Court took aim at the D.C. Circuit’s innovations
in Vermont Yankee. Then-Justice William Rehnquist, joined by a
unanimous court, strongly chastised the D.C. Circuit for having
“unjustifiably intruded into the administrative process” and imposed
338. See Barkow, supra note 64.
339. See, e.g., Daniel E. Ho, Congressional Agency Control: The Impact of Statutory Partisan
Requirements on Regulation (Feb. 12, 2008) (unpublished manuscript) (examining effects of
partisanship requirements on preferences of agency officials), http://dho.stanford.edu/research/
partisan.pdf.
340. See generally Bulman-Pozen, supra note 126.
341. See supra Part I.B; see also Reuel E. Schiller, Enlarging the Administrative Polity:
Administrative Law and the Changing Definition of Pluralism, 1945-1970, 53 VAND. L. REV. 1389
(2000) (discussing cultural changes that precipitated administrative law reforms at that time).
342. See Stewart, supra note 68, at 1805–07.
343. Bressman, supra note 67, at 475.
2 LIVERMORE 45-134 (DO NOT DELETE)
118
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
requirements that “border[] on the Kafkaesque.”344 The Court held that the
Administrative Procedure Act “establishe[s] the maximum procedural
requirements which Congress was willing to have the courts impose” and
implied that anything more allowed a court to invalidate a rule when it was
“unhappy with the results reached” by the agency. 345
Vermont Yankee garnered a mixed reception in the D.C. Circuit.346 On
the one hand, that court embraced the Vermont Yankee message in cases
like Sierra Club v. Costle that refused to find that White House influence
invalided a rulemaking process, stating that courts were not authorized to
“convert informal rulemaking into a rarified technocratic process,
unaffected by political considerations or the presence of Presidential
power.”347 On the other hand, some procedural limitations on informal
rulemaking—such as the Portland Cement doctrine that agencies must
disclose data and studies in their notices of proposed rulemakings—have
survived, and the Supreme Court has not seen fit to issue a follow-up
decision extending Vermont Yankee to void these requirements.348
Even had Vermont Yankee fully stopped the courts from imposing
additional procedural requirements on agencies, there remains a role for
probing substantive review of agency decisions for legality and reasoning.
In State Farm, the Court reiterated the legitimacy of hard look review,
leading to semi-procedural requirements on agencies to compile
voluminous administrative records and respond to substantive comments
received through the notice-and-comment process. Some commentators
have argued that these reason-giving requirements have, essentially,
replicated the most onerous procedural demands under a different name.349
344. Vt. Yankee Nuclear Power Corp. v. Natural Res. Def. Council, 435 U.S. 519, 556–58
(1978).
345. Id. at 524.
346. The D.C. Circuit’s treatment of ex-parte contacts requirements in informal rulemaking is
instructive. In two cases issued in 1977, the D.C. Circuit issued conflicting opinions on ex-parte contact
limitations in informal rulemaking. See Action for Children’s Television v. FCC, 564 F.2d 458 (D.C.
Cir. 1977) (declining to limit ex parte contacts); Home Box Office, Inc. v. FCC, 567 F.2d 9 (D.C. Cir.
1977) (limiting ex parte contacts). After Vermont Yankee, the foray into policing ex-parte contacts was
essentially abandoned. See, e.g., United Steelworkers v. Marshall, 647 F.2d 1189, 1214 (D.C. Cir.
1980) (recognizing that post-Vermont Yankee, if there is no statutory basis for a procedural protection,
“that is virtually the end of the inquiry,” except where the lack of a procedure “violated the due process
rights of the petitioners” or there are “extremely compelling circumstances” that allow courts to
“impose nonconstitutional extra-statutory procedures on agencies” (internal quotation marks omitted)).
347. Sierra Club v. Costle, 657 F.2d 298, 408 (D.C. Cir. 1981). For a critical take on Sierra Club
v. Costle, see Heinzerling, supra note 80 at 178–79 (saying that the decision turns rulemaking into
decision a “charade,” “undermines the fundamental rationale for administrative agencies,” and “perverts
the place of expertise in administrative law”).
348. See Beermann & Lawson, supra note 83.
349. See Pierce, supra note 83.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
119
Given the penetrating role of hard look review, it is worth considering
how this institutional arrangement fits with responsible party
administration, and in particular whether, and how, it promotes either party
responsibility or administrative values.
Judicial review under the arbitrary and capricious standard increases
the cost and length of the rulemaking process, but also helps ensure that
agencies examine the most important consequences of their regulatory
actions and consider public input.350 In the language of administrative
values, arbitrary and capricious review could be thought to privilege
expertise, and perhaps representativeness, over the vigor and coherence
brought by executive leadership.351 Party government is also affected by
arbitrary and capricious review. Slowing down the rulemaking process
makes it more difficult for incoming administrations to carry out their
party’s policy programs through regulation—it can takes months, and even
years, for a new administration to see its policy preferences carried through
the rulemaking process, and resource constraints limit the number of policy
domains that a new administration can influence. This lag reduces the
policy consequences of elections, undermining party government.352
If a different balance between administrative values and party
responsiveness is desirable, arbitrary and capricious review is a potential
candidate: more lenient review would enhance the ability of parties to
implement their programs; more probing review would facilitate at least
some of the administrative values (although at a cost to others). But such
reform must be scrutinized carefully. One proposal, debated in a recent
spate of law review articles, would allow agencies to provide “political
reasons” for their decisions. 353 Those who favor political reasons argue that
350. Compare Thomas O. McGarity, Some Thoughts on “Deossifying” the Rulemaking Process,
41 DUKE L.J. 1385 (1992) with Mark Seidenfeld, Why Agencies Act: A Reassessment of the Ossification
Critique of Judicial Review, 70 OHIO ST. L.J. 251 (2009) (defending judicial review against claims that
it imposes undo status quo bias on agencies).
351. Professor Mark Seidenfeld argues that the seeming tension between hard look review and
political influence is frequently overstated. See Mark Seidenfeld, The Irrelevance of Politics for
Arbitrary and Capricious Review, 90 WASH. U. L. REV. 141 (2012). Specifically, Seidenfeld argues that
agency expertise is used to clarify the tradeoffs between regulatory alternatives and policy motivations
inform the choice between those alternatives. Id. Hard look review, under this account, facilitates
expertise without necessarily undermining accountability to political leaders.
352. Even under Seidenfeld’s account of hard look review, id., arbitrary and capricious review,
by slowing down the regulatory process and injecting courts into regulatory decision making, reduces
the net influence of parties on administrative action.
353. See Enrique Armijo, Politics, Rulemaking, and Judicial Review: A Response to Professor
Watts, 62 ADMIN. L. REV. 573 (2010); Stephen M. Johnson, Disclosing the President’s Role in
Rulemaking: A Critique of the Reform Proposals, 60 CATH. U. L. REV. 1003 (2011); Scott A. Keller,
Depoliticizing Judicial Review of Agency Rulemaking, 84 WASH. L. REV. 419 (2009); Nina A.
Mendelson, Disclosing “Political” Oversight of Agency Decision Making, 108 MICH. L. REV. 1127
(2010); Jodi L. Short, The Political Turn in American Administrative Law: Power, Rationality, and
Reasons, 61 DUKE L.J. 1811 (2012); Glen Staszewski, Political Reasons, Deliberative Democracy, and
2 LIVERMORE 45-134 (DO NOT DELETE)
120
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
such a regime will better comport with the practical reality of political
control and would increase accountability to voters by clarifying lines of
authority between the President and administrative decisions.
But while political reasons would likely increase party responsiveness,
the responsible party administration framework may provide good reasons
to be skeptical. An expanded role for political reasons may shift the balance
of power toward irresponsible parties by making it more difficult for
centrist, technocratic-oriented constituencies within parties to resist
demands from the extremes.354 If agencies are required to give publicregarding reasons for their actions, it will be costlier and riskier to pursue
policies that have little connection to some plausible public interest.355 If
political reasons lower the costs of pursuing policies that are disconnected
from broad national interests, they may shift power within parties in favor
of intense policy demanders at the expense of those actors within parties
whose incentives better align with broad national interests. The net results
may be an increase in party government that comes at the expense of
administrative values (such as legality) as well as increased risk of party
irresponsibility.
Although the Court has given no indication that it intends to adopt
political reasons, it has also been hesitant to heighten the standard of
review to counteract risks specifically generated by party control over
agencies. Perhaps most obvious was the failure of the Court to adopt a
doctrine disfavoring inconsistent agency action in its 2009 decision Fox v.
FCC.356 A robust doctrine on consistency helps mitigate one of the inherent
risks associated with partisan oversight over agencies—policy oscillation.
When agencies change their positions to reflect an incoming
administration, the result may be new policies that no better reflect the
interests of voters but that impose costs by upsetting settled expectations
and creating the risk of future, unproductive policy uncertainty.357
Administrative Law, 97 IOWA L. REV. 849 (2012); Kathryn A. Watts, Proposing a Place for Politics in
Arbitrary and Capricious Review, 119 YALE L.J. 2 (2009).
354. Cf. Short, supra note 353, at 1869 (noting the potential for political reasons to shift authority
within the agency to shift between expert-oriented staff and politically minded appointees). See
generally Elizabeth Magill & Adrian Vermeule, Allocating Power within Agencies, 120 YALE L.J. 1032
(2011).
355. Under Seidenfeld’s characterization, supra note 351, requiring agencies to accompany their
regulatory choices with neutral explanations of the tradeoffs involved similarly makes it difficult for
agencies to be responsive to more extreme constituencies without the controlling party.
356. See FCC v. Fox Television Stations, Inc., 556 U.S. 502 (2009).
357. Concerns about policy oscillation have been a driving motivator in the creation of
independent agencies. Barkow, supra note 64, at 24. Policy stability is seen as particularly important in
the context of financial regulation. See Jacob E. Gersen, Designing Agencies, in RESEARCH HANDBOOK
ON PUBLIC CHOICE AND PUBLIC LAW 333, 348 (Daniel A. Farber & Anne Joseph O’Connell eds.,
2010); Devins and Lewis, supra note 62, at 465–66 (discussing the Fed); David E. Lewis, The Adverse
Consequences of the Politics of Agency Design for Presidential Management in the United States: The
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
121
Nevertheless, the Court held in FCC v. Fox that, although “[a]n agency
may not . . . depart from a prior policy sub silentio [,] . . . it need not
demonstrate to a court’s satisfaction that the reasons for the new policy are
better than the reasons for the old one.” 358 Despite the outcome in FCC v.
Fox, the Court’s hesitation to develop an explicit doctrine on consistency
garnered only a bare majority, and perhaps not even that, given a
concurrence by swing-vote Justice Kennedy that agreed with portions of
Justice Breyer’s dissent.359 It is possible that future decisions will increase
the cost of inconsistency, changing the calculus for political parties toward
focusing on establishing, rather than changing, agency positions. Arguably,
such an arrangement would leave adequate room for party government
while better protecting administrative values.
The Chevron doctrine also has complicated consequences for
responsible party administration. In Chevron, the Court reaches in many
different normative directions to justify judicial deference to agency
interpretations, which at least arguably conflicts with the value of
legality.360 In practice, the consequences of deference for administrative
values is often contingent on party control of the executive, a factor that the
Court does not (explicitly) consider. For example, when the President’s
preferences are aligned with the mission of an agency, then the value of
expertise is likely to be well served by deference because the policy
choices embodied in regulation will be more likely to reflect the views of
the expert bureaucrats at an agency. On the other hand, when the policy
program of the party in control is out of sync with an agency’s mission,
then deference will ill-serve neutral competence; the outcome selected by
political appointees may well differ from the judgment of the career civil
servants.361 Exactly the opposite consequences arise for representativeness.
Relative Durability of Insulated Agencies, 34 BRIT. J. POL. SCI. 377, 400 (2004) (noting that “[i]n
insulated agencies the impact of changing administrations is muted so that policies have less variance”);
Vermeule, supra note 63.
358. Fox, 556 U.S. at 515.
359. Id. at 535–39
360. Chevron, U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 865 (1984). The
Court notes its relative incompetence in making judgments in technically complex areas of law. See id.
at 865 (“Judges are not experts in the field . . . .”). The Court also raises a representativeness argument,
noting that agencies, through the President, are “directly accountable to the people.” Id. at 864–66. The
Court also grounds deference in the value of executive leadership, finding that it is appropriate for an
agency to “rely upon the incumbent administration’s views of wise policy to inform its judgments.” Id.
In addition, an alternative “official theory” has developed that Chevron reflects adherence to the intent
of Congress. See Adrian Vermeule, No, 93 TEX. L. REV. 1547, 1555–57 (2015) (reviewing PHILIP
HAMBURGER, IS ADMINISTRATIVE LAW UNLAWFUL? (2014)) (reciting, without endorsing, the
delegation theory of Chevron).
361. See Jody Freeman & Adrian Vermeule, Massachusetts v. EPA: From Politics to Expertise,
2007 SUP. CT. REV. 51 (arguing that the Court, in Massachusetts, was concerned that political
considerations overrode agency expertise).
2 LIVERMORE 45-134 (DO NOT DELETE)
122
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
When the party in control of the White House is in conflict with an
agency’s mission, that tension may reduce the risk of a non-representative
outcome, and deference serves its goal.362 When the party in control of the
White House is aligned with an agency’s mission, there is greater risk of an
outcome that tracks partisan, rather than majoritarian, preferences.
Although Chevron deference has ambiguous consequences for the
administrative values at its putative foundations, from the perspective of
party government, deference clearly increases the ability of parties to
implement their programs after electoral success.363 The Court, however,
may have recently signaled an appetite for scaling back deference in
exactly the context where it most facilitates responsible party government.
In King v. Burwell, upholding health care subsidies in the Affordable Care
Act, Chief Justice Roberts, writing for a six Justice majority, included the
following language:
When analyzing an agency’s interpretation of a statute, we often
apply the two-step framework announced in Chevron. Under that
framework, we ask whether the statute is ambiguous and, if so,
whether the agency’s interpretation is reasonable. This approach
“is premised on the theory that a statute’s ambiguity constitutes an
implicit delegation from Congress to the agency to fill in the
statutory gaps.” “In extraordinary cases, however, there may be
reason to hesitate before concluding that Congress has intended
such an implicit delegation.”364
The Court then went on to examine the relevant statutory issue, finding that
it was a “question of deep economic and political significance” and
therefore, “had Congress wished to assign that question to an agency, it
surely would have done so expressly.”365
The Court’s move is problematic for reasons offered by Justice Breyer
in his dissent in Brown & Williamson, the case that Justice Roberts relies
on in Burwell.366 It is exactly the high-profile questions where party
362. This effect could be analogized to so-called panel effects on appellate courts, a phenomenon
in which panels that draw judges with different partisan affiliations appear to reach systematically
different conclusions than panels that draw judges with affiliations of only one party. See generally
VIRGINIA A. HETTINGER, STEFANIE A. LINDQUIST & WENDY L. MARTINEK, JUDGING ON A COLLEGIAL
COURT (2006).
363. Cf. Sunstein, supra note 28.
364. King v. Burwell, 135 S. Ct. 2480, 2488–89 (2015) (citations omitted).
365. Id. at 2489.
366. See FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 190–91(2000) (“[T]he very
importance of the decision taken here, as well as its attendant publicity, means that the public is likely
to be aware of it and to hold those officials politically accountable. Presidents, just like Members of
Congress, are elected by the public. Indeed, the President and Vice President are the only public
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
123
influence over agency decision making is most likely to be responsible—
voters are paying attention, and if they are displeased, they will know who
to blame. The opposing party has plenty of incentives to highlight
difference on those high-profile issues and make its case that the program
of the party in power, as developed in the regulatory action at issue, is
undesirable. When the electoral consequences of agency decisions are
substantial, party responsibility is at its zenith, and the legitimate scope of
party influence over agency decisions making is greatest. The Court in
Burwell turns this logic on its head, seemingly indicating that the decisions
of less public concern, in which voter inattention is greatest, and the risks
of party pandering to intense policy demanders is greatest, are those that
deserve the greatest deference.
It bears mentioning generally in the context of judicial review that
courts themselves are not outside the influence of contemporary parties.
Although the role of partisan commitments in affecting judicial decision
making can be overstated, moral or ideological commitments do play a role
and, even assuming good faith behavior on the part of judges, presidents,
and senators, when making appointment and confirmation decisions, can
“take into account judicial characteristics that may potentially bear on the
moral and political questions that appropriately arise in legal disputes.”367
Given the role of the political branches in judicial appointments, it should
come as no surprise that the rise of contemporary parties has, in fact, been
associated with an increasingly contentious judicial confirmation
process.368 Judicial review of agency action falls into a larger set of issues
implicated by the potential for contemporary parties to negatively affect the
functioning of the courts. In particular, the incorporation of regulatory
programs into national politics increase the payoff of using the nomination
and confirmation process to screen judicial candidates based on signals
related to policy attitudes, with potentially negative effects for the
independence and functioning of the courts. Proposals to “scrap[]” the
presumption of judicial reviewability of agency action or scale back the
role of courts in examining agency structure may help reduce the threat of
an increasingly polarized judiciary.369 Such proposals, of course,
officials whom the entire Nation elects. I do not believe that an administrative agency decision of this
magnitude—one that is important, conspicuous, and controversial—can escape the kind of public
scrutiny that is essential in any democracy.”).
367. Harry T. Edwards & Michael A. Livermore, Pitfalls of Empirical Studies that Attempt to
Understand the Factors Affecting Appellate Decisionmaking, 58 DUKE L.J. 1897, 1943 (2009).
368. See O’Connell, supra note 26.
369. Nicholas Bagley, The Puzzling Presumption of Reviewability, 127 HARV. L. REV. 1285,
1335 (2014) (“Because it is unjustified in principle and harmful in practice, the presumption of
reviewability should be scrapped.”); Huq, supra note 17 (proposing a less extensive role for the courts
in supervising agency structure). See also Cass R. Sunstein & Thomas J. Miles, Depoliticizing
2 LIVERMORE 45-134 (DO NOT DELETE)
124
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
necessarily risk sacrificing the administrative values that are promoted by
judicial review.370 Ultimately, in a political environment that is dominated
by contemporary political parties, the degree of and risk for polarization of
the judiciary are among the factors that must be weighed against competing
considerations when evaluating the normative desirability of the current
regime of a relatively strong role for courts in overseeing administrative
agencies.
C. Executive Review
Before adopting any major substantive regulation, agencies are
required by executive orders—in place since the Reagan administration—
to submit their proposal for review by the Office of Information and
Regulatory Affairs in the White House. This review process is often held
out as the archetypical example of presidential administration.371 Presidents
from both political parties have endorsed regulatory review, and it is now
an enduring feature of the American regulatory state. Nevertheless,
regulatory review remains deeply controversial, and proposals to reduce its
influence or eliminate the practice altogether are periodically floated.372
Given its prominent role, and the consistent criticism that it has faced, it is
useful to examine how well the institution of regulatory review conforms to
the normative goals of responsible party administration by facilitating
either party responsibility or administrative values.
Perhaps surprisingly, given its important policy role, OIRA is staffed
almost entirely by civil service bureaucrats, rather than political appointees.
These bureaucrats are typically trained in economics, statistics, and public
policy; many of them do not turn over with a change in party control.373
Only the Administrator and two special assistants are political
appointees.374 As a consequence of this arrangement, it is common for the
work of political appointees at agencies to be evaluated by career
Administrative Law, 58 DUKE L.J. 2193 (2009) (examining partisan voting in administrative law
decisions and proposing potential remedies, including mixed party panels).
370. See Bagley, supra note 369, at 1330 (presumption of reviewability “could foster adherence
to law, improve agency deliberation, and increase the accuracy of agency decisions”). Purely from the
perspective of responsible party government, partisanship on the courts has few benefits. Although it
may increase the number of decisions subject to party influence, the lag between electoral victory and
policy outcomes can be enormous. The beneficial feedback loop between elections and policy
anticipated by responsible party government is effectively short-circuited by life tenure.
371. See, e.g., Kagan, supra note 3, at 2277–78.
372. See Steinzor, supra note 330.
373. See William F. West, The Institutionalization of Regulatory Review: Organizational
Stability and Responsive Competence at OIRA, 35 PRESIDENTIAL STUD. Q. 76, 84 (2005).
374. Id.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
125
bureaucrats at OIRA.375 Even for the political appointees, there is a wellestablished convention for appointees to be drawn from a pool of
technocratic experts who do not have strong ties to particular interest
groups.376
But although, at first glance, OIRA review might seem inconsistent
with party control, a prominent role for career bureaucrats in regulatory
oversight may, perhaps paradoxically, be well suited to conforming agency
regulatory output to party programs. Contemporary party programs are
policy slates that represent compromise between various constituencies.
Securing and deploying the resources needed to gain and exercise power
depends on developing programs that gather support from a sufficiently
broad group of constituencies.377 Although the constituencies within parties
work cooperatively to gain advantage against the competing party, they
also work competitively to maximize their own advantage vis-à-vis each
other in securing the benefits of party control. Especially when the program
is being applied in specific instances during the course of a presidential
administration, mediating conflict between disparate constituencies is
essential.
Single-issue agencies may be poorly suited to navigate this intraparty,
inter-constituency terrain. While in power, the collective good of the party
coalition is served through the production of some optimal set of regulatory
decisions, subject to legal and political constraints. In a complex policy
environment, rarely will a regulatory choice simply provide a benefit to one
constituency without any offsetting cost to another or reduction in electoral
potential (which harms all party constituencies). Single-issue agencies may
lack the perspective necessary to properly balance these competing
concerns, and may be subject to overbearing influence by particular interest
groups. In effect, there are risks that individual agencies may be “captured”
by specific constituencies in ways that are detrimental to the aggregate
good of the party in power.378
375. Livermore, supra note 132, at 622–23. This arrangement inverts the standard structure of
political-appointee supervision of career bureaucrats. The model of OIRA review can be contrasted
with a hypothetical review process run through some other office, such as the Domestic Policy Council,
that was staffed primarily with political appointees rather than careerists. Indeed, at the state level,
regulatory review is sometimes overseen by much more overtly political institutions. See id. at 685
n.364 (discussing Executive Order of Andrew Cuomo, transferring regulatory review power to overtly
political office). Cf. West, supra note 373, at 86–89 (arguing that OIRA’s career staff subvert whatever
policy preferences they have to respond to the needs of the political principals in power).
376. Livermore & Revesz, supra note 34, at 1373.
377. See supra Part III.B.
378. On agency capture, see generally PREVENTING REGULATORY CAPTURE: SPECIAL INTEREST
INFLUENCE AND HOW TO LIMIT IT (Daniel Carpenter & David A. Moss eds., 2014).
2 LIVERMORE 45-134 (DO NOT DELETE)
126
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
OIRA review may help reduce this capture risk.379 As a generalist
review institution, without strong ties to any particular constituency, there
is less potential for overbearing influence by any single interest group.
Because OIRA review touches on a wide variety of issue areas and is
carried out by career bureaucrats, incentives to invest in influence are weak
and highly diffuse. OIRA review also plays a strong coordinating role,
soliciting views across the federal government on the regulatory proposals
of each agency. By initiating interagency comment and acting as a
clearinghouse for disparate views, OIRA review enhances the probability
that the potential downsides of regulatory proposals will be aired and
vetted. In essence, OIRA review helps identify those issues where conflicts
between party constituencies are likely to arise and helps facilitate
deliberation over how to settle those conflicts. In this way, OIRA review
helps limit the risk of irresponsible party government by reducing the risk
of intense policy demanders exploiting single-issue agencies and thereby
moving the party away from majority preferences.
At the same time that OIRA may help facilitate responsible party
government, its role in promoting administrative values is less clear.
OIRA’s location within the White House and ability to balance competing
intraparty demands may allow it to insert technocratic criteria into political
oversight.380 But OIRA has also been criticized for lacking substance
expertise, and, as a practical matter, the breadth of the issues that OIRA
staff must deal with, and the tight deadlines in which much of the work is
accomplished,381 may preclude building deep substantive expertise in
particular issue areas. Although OIRA may be thought to have had a certain
kind of expertise, in regulation more generally, agency officials have
comparative strengths in issue-specific expertise. OIRA’s coordinating
function may help bring some measure of coherence to an administration’s
regulatory agenda, but OIRA review has also been criticized for
contributing to regulatory “ossification” and reducing the vigor and
responsiveness of agencies.382 This problem is especially troubling because
OIRA review is focused on agency action, while the failure to act is not
379. The logic in this and the following two paragraphs tracks Livermore & Revesz, supra note
34, at 1361–73.
380. Livermore & Revesz, supra note 34.
381. Alex Bolton, Rachel Augustine Potter & Sharece Thrower, Presidential Oversight and
Regulatory Delay: How Politics and Organizational Capacity Influence OIRA Rule Review
(unpublished
manuscript)
http://www.lsa.umich.edu/UMICH/polisci/Home/Events/Thrower,%20
Presidential%20Oversight%20and%20Regulatory%20Delay.pdf (examining effect of staffing shortfalls
on the pace of executive review).
382. See McGarity, supra note 350.
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
127
subject to similar scrutiny.383 From the perspective of representativeness,
OIRA plays a more ambiguous role. Review may reduce the risk of agency
capture, but OIRA itself is not subject to public participation and
transparency requirements of the Administrative Procedure Act, and critics
have argued that participation in the OIRA review process by outsiders
tends to skew in favor of regulated industry.384
An area where OIRA has come under particular criticism for
undermining administrative values is the issue of delay, and in particular
delay that appears to be related to the electoral calendar. In addition to the
typical group of critics who are skeptical of regulatory review in general,
the Administrative Conference of the United States recently made several
recommendations to OIRA to reduce undue delays in regulatory review.385
Some have argued that during the Obama administration, controversial
rules were put on hold in the months leading up to the 2012 presidential
election.386 Especially for agencies subject to court orders or statutory
deadlines, these electorally motivated delays undermine the value of
legality in the administrative process.
OIRA has also been criticized for inserting extra-statutory
considerations into agency rulemaking, in particular through the
requirement that agencies conduct cost-benefit analysis of their major
rulemakings.387 There is a lively debate on this question. Some
commentators have argued for a broad background norm that allows
agencies to conduct some kind of weighing of costs and benefits, absent a
clear statutory directive to the contrary.388 Others have strenuously
objected, arguing, in effect, for the opposite default—absent a clear
directive to consider costs, agencies should be prohibited from doing so.389
In cases such as EME Homer and Entergy v. Riverkeeper (both interpreting
environmental provisions that are silent on cost considerations) the Court
383. Livermore & Revesz, supra note 34.This bias against agency action may generate a partisan
bias in favor of Republican interests, which generally favor less stringent regulation.
384. RENA STEINZOR ET AL., CTR. FOR PROGRESSIVE REFORM, WHITE PAPER 1111, BEHIND
CLOSED DOORS AT THE WHITE HOUSE: HOW POLITICS TRUMPS PROTECTION OF PUBLIC HEALTH,
WORKER SAFETY, AND THE ENVIRONMENT (2011) (documenting skewed participation in OIRA review
process, with representatives from regulated industry participating at higher rates).
385. ADMINISTRATIVE CONFERENCE OF THE UNITED STATES, ADMINISTRATIVE CONFERENCE
STATEMENT NO. 18, IMPROVING THE TIMELINESS OF OIRA REGULATORY REVIEW (Dec. 6, 2013).
386. See id. at 4 (“Senior agency employees provided a variety of perspectives as to why they
believe that OIRA review times increased in 2012–13, including . . . concerns by some in the Executive
Office of the President (EOP) about the issuance of potentially costly or otherwise controversial rules
during an election year . . . .”).
387. See, e.g., Heinzerling, supra note 181.
388. See, e.g., Cass R. Sunstein, Cost-Benefit Default Principles, 99 MICH. L. REV. 1651 (2001).
389. See, e.g., BETTER MARKETS, SETTING THE RECORD STRAIGHT ON COST-BENEFIT ANALYSIS
AND FINANCIAL REFORM AT THE SEC 28–32 (2012).
2 LIVERMORE 45-134 (DO NOT DELETE)
128
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
appears to lean toward the former interpretation, preserving agency
discretion to consider costs when statutes are unclear.390 As a practical
matter, even when statutorily prohibited from considering costs—such as
when the EPA sets the national ambient air quality standards—it appears
that agencies nonetheless weigh the downsides of regulation.391 Still, to the
extent that the anti-cost-consideration camp has the better legal argument,
OIRA’s role in encouraging agencies to engage in cost-benefit analysis
could run counter to their statutory authority.
Against this arguable tradeoff against the administrative value of
legality are, potentially, important benefits for responsible party
government that may be brought about by the systematic use of cost-benefit
analysis. Most generally, the comprehensive nature of the technique
accords well with a role for OIRA in helping avoid irresponsible party
government. Agencies, operating under specific statutory directives and
oriented toward particular policy areas may be subject to myopia, giving
undue attention to particular classes of interests while ignoring or
underemphasizing others. Cost-benefit analysis can help illuminate a
broader range of regulatory effects that matter for a diverse set of party
constituencies.
The methodology of cost-benefit analysis is also, itself, a site of
partisan contestation. When OIRA review began, the methodology of
regulatory cost-benefit analysis was in a stage of relative infancy. There
were basic techniques for estimating regulatory effects and assigning rough
dollar values, but standardized methods that could guide agencies or OIRA
in carrying out or evaluating these analyses did not exist.392 Over time, the
methodology of regulatory cost-benefit analysis was constructed,
essentially from whole cloth, in ways that promote some interests at the
expense of others.393
390. EPA v. EME Homer City Generation, L.P., 134 S. Ct. 1584 (2014) (upholding EPA rule on
interstate air pollution in which the agency considered costs when setting emissions allowances for
states); Entergy Corp. v. Riverkeeper, Inc., 556 U.S. 208 (2009) (deferring to agency’s interpretation of
Clean Water Act regulations to allow for a cost-based variance procedure). See also Michigan v. EPA,
135 S. Ct. 2699 (2015) (requiring agency to consider costs when adopting hazardous air pollution
standards for power plants).
391. See Michael A. Livermore & Richard L. Revesz, Rethinking Health Based Environmental
Standards, 89 N.Y.U. L. REV. 1184, 1231–33 (2014).
392. The first guidelines issued by the White House on conducting regulatory impact analysis
were highly perfunctory, and there was some time before significant centralized guidance was provided.
Livermore, supra note 132, at 640–46.
393. See id. (noting that the EPA’s preferred methodology often carried the day). As carried out,
cost-benefit analysis relies heavily on welfare economics for its intellectual underpinning, but that is a
choice itself, and there are alternatives, such as happiness analysis. See generally JOHN BRONSTEEN,
CHRISTOPHER BUCCAFUSCO & JONATHAN S. MASUR, HAPPINESS AND THE LAW (2014) (proposing
happiness analysis as an alternative to cost-benefit analysis).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
129
Initially, support for cost-benefit analysis split along partisan lines,
with Democrats and affiliated interest groups strongly opposing Reagan’s
move to elevate the technique.394 Over time, bipartisan support for the
technique grew, and partisan debates began to focus on methodological
questions, including: the discount rate for future costs and benefits; the
monetary value of reducing statistically small mortality risks; whether lives
or life expectancy is the correct metric for mortality risk reduction; the
default dose-response relationships to use for different classes of
environmental pollutants; the effect of mandatory energy efficiency
requirements on consumer welfare; the value of overseas climate change
damages; the types of models to be used for estimating employment effects
of regulation; and the relationships between corporate disclosure and
systematic risk in the financial sector.
Positions on these questions can be, and to some extent have been,
incorporated into party programs. Researchers Art Fraas and Richard
Morgenstern examined fifteen years of statutorily required annual reports
prepared by OIRA on the costs and benefits of federal rulemakings.395
Fraas and Morgenstern examine and compare the analytic priorities in
Democratic and Republican administrations, finding important crossadministration differences. For example, there were differences in the
“relative emphasis on difficult-to-measure” variables, with the Obama
administration focused on regulatory benefits, such as the “value of
ecological services and the value of dignity and equity,” while during the
Bush administration, there was greater “discussion of the uncounted costs
of regulation, including those faced by small businesses.”396 Similar
differences are found in the treatment of discounting,397 and in
unemployment.398 Other indications that cost-benefit analysis methodology
has found its way into party programs include 2012 Republican presidential
394. See generally RICHARD L. REVESZ & MICHAEL A. LIVERMORE, RETAKING RATIONALITY:
HOW COST-BENEFIT ANALYSIS CAN BETTER PROTECT THE ENVIRONMENT AND OUR HEALTH (2008).
See, e.g., Philip Shabecoff, Reagan Order on Cost-Benefit Analysis Stirs Economic and Political
Debate, N.Y. TIMES, Nov. 6, 1981, at 28 (quoting Democratic Congressman Henry A. Waxman, among
others, in opposition to the technique).
395. See Art Fraas & Richard Morgenstern, Identifying the Analytical Implications of Alternative
Regulatory Philosophies, 5 J. BENEFIT COST ANALYSIS 137, 139 (2014).
396. Id. at 141.
397. Id. at 142 (Obama administration emphasizing future benefits).
398. Id. (“[T]he Bush administration tends to emphasize the trade-offs between regulations and
more broadly defined economic performance” while the Obama administration focuses on “those
analyses that find negligible impacts or even a positive relationship between regulation and
employment.”).
2 LIVERMORE 45-134 (DO NOT DELETE)
130
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
candidate Mitt Romney’s endorsement of jobs impact analysis and the
partisan character of recent debates on the social cost of carbon.399
Although the parties differ on certain methodological questions in costbenefit analysis, it is important to note the degree of overlap and
similarity.400 There are three general constraints that pull parties toward a
fairly similar interpretation of cost-benefit analysis. The first is the need to
balance the demands of relatively extreme elements within the party
against the desire to appeal to a sufficiently broad swath of the electorate.
Although very few voters have preferences over cost-benefit analysis
methodologies, party constituencies care about the consequences that
analytic choices have for regulatory outcomes.401 Extreme positions on
methodological questions will be difficult to maintain in the face of the
general need to balance the demands of different groups alongside mass
appeal.402
Second, cost-benefit analysis also provides technocratic legitimation to
OIRA oversight, which could be eroded by unusual or extreme
methodological choices. Despite the novelty of its application to regulation,
at the time of the Reagan order, cost-benefit analysis already had a
substantial pedigree in other settings.403 Defenders of Reagan’s innovation
could draw on this pedigree when seeking to legitimate regulatory
review.404 Were the practice of cost-benefit analysis to depart too radically
from the norms of the economic profession on which it was ostensibly
399. See Michael A. Livermore & Richard L. Revesz, Interest Groups and Environmental
Policy: Inconsistent Positions and Missed Opportunities, 45 ENVTL. L. 1, 9 (quoting Mitt Romney
campaign documents); see also Zack Colman, GOP Senators Slam ‘Significant Change’ to Carbon
Costs THE HILL (June 19, 2013, 9:06 PM), http://thehill.com/policy/energy-environment/306645-gopsenators-slam-significant-change-to-social-cost-of-carbon.
400. Fraas & Morgenstern, supra note 395, at 169 (“The cross-administration differences we
identify appear to reflect relatively modest shifting across political parties on issues where reasonable
people may disagree.”).
401. For example, a higher social cost of carbon will favor more stringent greenhouse gas
regulation.
402. See supra Part III.B.
403. See Jonathan B. Wiener, The Diffusion of Regulatory Oversight, in THE GLOBALIZATION OF
COST-BENEFIT ANALYSIS IN ENVIRONMENTAL POLICY 123, 134 (Michael A. Livermore & Richard L.
Revesz, eds., 2013) (discussing roots of cost-benefit analyses in the French civil engineering tradition
that spanned back nearly a century); see Edward P. Fuchs & James E. Anderson, The
Institutionalization of Cost-Benefit Analysis, 10 PUB. PRODUCTIVITY REV. 25, 25 (1987) (discussing use
of cost-benefit analysis in Robert McNamara’s Department of Defense in the 1960s).
404. See Shabecoff, supra note 394 (stating that defenders of cost-benefit analysis at the time of
the Reagan order noted that the technique has been used “for many years [by] the Army Corps of
Engineers . . . to justify building big dams and other costly projects”); Joseph Cooper & William F.
West, Presidential Power and Republican Government: The Theory and Practice of OMB Review of
Agency Rules, 50 J. POL. 864, 872 (1988) (arguing that cost-benefit analysis could “legitimate
presidential power” by helping to “dispel” concerns about presidential arbitrariness); DeMuth &
Ginsburg, supra note 182, at 1082 (arguing that the cost-benefit “standard and the OMB review
procedure” are “complementary” to each other).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
131
based, the technocratic legitimation for presidential review would lose its
force.
There are also more purely bureaucratic reasons that invest cost-benefit
analysis methodologies with inertia. Professor Sunstein refers to settled
methodological practices (and the documents that embody them) as “a kind
of common law for cost-benefit analysis” that is only changed after “an
extended process, which . . . involve[s] many officials and sometimes a
public comment period, and which is likely to bear fruit only if and when a
consensus emerges.”405 In addition, OIRA is not invested with final
authority to disapprove agency rules—when conflicts between an agency
and OIRA cannot be resolved, they are “elevated” to senior political
officials in the White House, and if serious enough, the President.406
Established cost-benefit analysis methodologies help minimize that
outcome by setting the standard norms that can be applied to multiple
rulemakings.407
Finally, the cost-benefit standard empowers agencies to have
substantive influence on the shape of regulatory review. 408 Over the past
three decades, it has largely fallen on agencies to develop the methodology
of cost-benefit analysis, based on their comparative advantage in technical
expertise and resources.409 This agency influence occurs over the course of
many years, and especially when arising from career bureaucrats, is a
consistent pressure across administrations from both parties.
Taken together, these considerations constrain the range of debate over
cost-benefit analysis methodology, and likely result in OIRA review
exerting a moderating influence on agency decision making. Of course, if a
President decides to depart from norms of economic efficiency in a
particular regulatory decision, the qualms of an OIRA desk officer are
unlikely to be a major roadblock.410 But other factors being equal, the
structure of OIRA review likely empowers more moderate political actors
within an administration at the expense of the wings of both parties.
Regulatory review has become perhaps the central force shaping how
party administration is achieved in practice by virtue of its role in balancing
technocratic considerations and politics, agency independence and White
House authority, and statutory directives and social considerations. OIRA’s
405. CASS R. SUNSTEIN, VALUING LIFE: HUMANIZING THE REGULATORY STATE 63 (2014).
406. See Cass R. Sunstein, The Office of Information and Regulatory Affairs: Myths and
Realities, 126 HARV. L. REV. 1838, 1856–58 (2013) (describing the process that occurs when
controversies between agencies and OIRA cannot be resolved at the staff level).
407. Livermore, supra note 132.
408. Id.
409. Id.
410. Stuart Shapiro, Unequal Partners: Cost–Benefit Analysis and Executive Review of
Regulations, 35 ENVTL. L. REP. 10433, 10438–39 (2005).
2 LIVERMORE 45-134 (DO NOT DELETE)
132
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
critics believe, in essence, that the current equilibrium is all wrong, arguing
alternatively that technocrats have run wild, or that politics has infected
regulatory decision making; that too much power rests in the White House,
or that agencies should be reined in; that too much weight is placed on
costs and benefits, or that there is a myopic focus on narrow statutory
goals. Behind these criticisms are, in essence, competing normative visions
of the meaning of responsible party administration that rest on differing
views concerning the contribution of party competition to democratic
politics, the current level of responsibility in party government, and the
importance of one or the other administrative values. Given the basic
tensions built into the concept of responsible party administration, these
debates are unlikely to be resolved in the foreseeable future; at least so long
as the administrative values that are deeply engrained in our institutional
culture survive, and contemporary political parties take something like their
current form.
CONCLUSION
Contemporary parties are now fully established on the political
landscape. When presidents come into office, they do so at the head of a
political party, having outlived intraparty challengers and led a successful
electoral contest against the opposing party. In a very short period of time,
they are called on to transition from no-holds-barred partisan campaigning
to the highly constrained task of governing.411 They must fill several
thousand appointed positions, balancing demands from campaign staff,
donors, and interest groups. They confront an unruly group of co-partisans
in Congress as well as a hostile opposition party. All of these features of
the contemporary political ecosystem can have profound consequences for
administration, political supervision of administrative agencies, and
administrative law. Along with reorganizations within the Executive
Branch and changes in judicial doctrine concerning the exercise of
presidential power over agencies, the revival of political parties in their
contemporary form fundamentally shape presidential administration. The
normative and practical consequences of presidential oversight cannot be
understood absent an accounting of contemporary parties.
This Article explores the consequences of contemporary political
parties for the efficacy of presidential supervision of the regulatory state;
the normative justification for presidential power over agencies; and the
411. Mario Cuomo is credited with the observation that “[y]ou campaign in poetry[;] [y]ou
govern in prose.” OXFORD DICTIONARY OF QUOTATIONS 247 (Elizabeth Knowles ed., 7th ed. 2009)
(first appearing in The New Republic (Apr. 8, 1985)).
2 LIVERMORE 45-134 (DO NOT DELETE)
2015]
Political Parties and Presidential Oversight
11/5/2015 9:20 AM
133
dynamic of shared oversight between the President and Congress. The
weakening of traditional party organizations and the rise of contemporary
parties that are comparatively professionalized, nationally oriented, and
programmatic likely enhance the ability of the White House to subject
agencies to presidential control. At the same time, the tendency of
contemporary party programs to drift toward the extreme preferences of
intense policy demanders at their base undermines—or at least
complicates—traditional justifications for presidential control. Rather than
relying on the President’s presumed fidelity to median-voter preferences,
proponents of presidential administration may be better served by relying
on the attractive characteristics of responsible party government as their
normative touchstone. Similarly, whether interactions with Congress
devolve into mere blood sport or can instead lead to democracy-enhancing
dialogue depends on whether congressional parties are able to maintain the
moderation necessary to hew to a responsible party government model.
This Article also examines the consequences of contemporary political
parties for administrative law values and proposes the concept of
responsible party administration to evaluate how well judicial and
executive institutions promote normatively desirable party control while
preserving important administrative law values. Because the influence of
contemporary parties over presidential oversight is pervasive, the concept
of responsible party administration can be applied to a host of
administrative law questions, from judicial doctrine to agency design. This
Article explores two salient and important areas: doctrines concerning
judicial deference to agency decisions, and the institution of OIRA review.
On the deference front, the Court appears to strike an odd posture—hesitant
to exert influence over areas where a stronger judicial role might promote
responsible party administration (e.g., to avoid oscillating policies), while
growing more assertive in areas where the judiciary may bring little benefit
(e.g., on question of major social importance). With respect to OIRA, this
Article argues that the institution of regulatory review plays a major role in
determining whether responsible party administration is carried out in
practice, and discusses some of the benefits and costs associated with the
current system of review.
For better or worse, contemporary political parties now structure
American political life. But just as they are unlikely to evaporate from the
political scene, they are unlikely to stay static as social and cultural forces
press on the parties—in their mutual interlocking conflict—to adapt. As
they evolve, parties will continue to demand a response from administrative
law scholars and from courts. Balancing the potential benefits that
contemporary parties may bring for democratic accountability against their
costs to values such as expertise, neutrality, accountability, and efficacy
2 LIVERMORE 45-134 (DO NOT DELETE)
134
11/5/2015 9:20 AM
Alabama Law Review
[Vol. 67:1:45
will likely continue to be one of the defining challenges for administrative
law in the early twenty-first century.